Vous êtes sur la page 1sur 314

Janet Carr

Roberta Shepherd
Stroke Rehabilitation
For Butterworth Heinemann:
Senior Commissioning Editor: Heidi Allen
Project Development Editor: Robert Edwards
Project Manager: Jane DingwalllPat Miller
Design Direction: George Ajayi
Guidelil1es for Exercise and Training
to Optimize Motor Skill
Janet H Carr EdD FACP
Honorary Associate Professor,
School of Physiotherapy,
Faculty of Health Sciences,
The University of Sydney, Australia
Roberta B Shepherd EdD FACP
Foundation Professor of Physiotherapy,
Honorary Professor,
School of Physiotherapy,
Faculty of Health Sciences,
The University of Sydney, Australia
An imprint of Elsevier Science Limited
2003, Elsevier Science Limited. All rights reserved.
The right of Janet Carr and Roberta Shepherd to be identified as authors
of this work has been asserted by them in accordance with the Copyright,
Designs and Patents Act 1988
No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without either the
prior permission of the publishers (Permissions Manager, Elsevier Science
Limited, Robert Stevenson House, 1-3 Baxter's Place, Leith Walk,
Edinburgh EH1 3AF), or a licence permitting restricted copying in the
United Kingdom issued by the Copyright Licensing Agency, 90 Tottenham
Court Road, London WIT 4LP.
First edition 2003
ISBN 0 7506 4712 4
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
A catalog record for this book is available from the Library of Congress
Medical knowledge is constantly changing. As new information becomes
available, changes in treatment, procedures, equipment and the use of
drugs become necessary. The authors and the publishers have taken care
to ensure that the information given in this text is accurate and up to date.
However, readers are strongly advised to confirm that the information,
especially with regard to drug usage, complies with the latest legislation
and standards of practice.
your source for books,
journals and multimedia
in the health sciences
policy is to use
paper manufactured
from sustainable forests
Printed in China by RDC Group Limited I
Preface VB
Acknowledgements IX
Section 1 Introduction 1
1 Brain reorganization, the rehabilitation environment,
measuring outcomes 3
Section 2 Training guidelines 33
2 Balance 35
3 Walking 76
4 Standing up and sitting down 129
5 Reaching and manipulation 159
Section 3 Appendices 207
6 Impairments and adaptations 209
7 Strength training and physical conditioning 233
8 Overview 259
References 267
Index 293
In this book we have set down guidelines for training critical motor actions
after stroke: walking, reaching and manipulation, balancing in sitting and
standing, and standing up and sitting down. The guidelines are designed to opti
mize motor performance, in other words, skill. They are science-based in that
we have preceded each set of guidelines with a brief resume of scientific findings
relevant to the action, including a description of major biomechanical charac
teristics, and of muscle activity, and the changes that occur in performance due
to impairments and adaptations. Since stroke is the model used, details of age
related changes in performance are also provided. The guidelines themselves
include methods of task-oriented training of motor control to maximize skill,
environmental manipulations to foster cognitive engagement and enhance skill
learning, and methods of increasing muscle strength, soft tissue flexibility,
endurance and fitness. A short list of appropriate methods of measurement is
included. The guidelines are based on the limited evidence so far available from
clinical outcome studies, and each training chapter includes a table of recent
clinical trials we consider are of particular clinical interest, trials in which the
interventions are clearly outlined. The mechanisms of impairments and adapt
ations and their impact on performance, the task-dependence of strength training
and conditioning, and a brief overview of the role of physiotherapy in the early
care of individuals who have had a stroke, are included in appendices.
Bridging the gap between science and practice is an overwhelming task for
the clinician, who must endeavour to keep up to date in scientific findings in a
wide range of subjects, as well as work out how to use the information to inter
vene with clinical problems. Illustrating ways of bridging the gap has therefore
been a critical aim for us in writing textbooks over the last two decades. It is 20
years since we published the first edition of A Motor Relearning Programme
for Stroke, with an updated version in 1987. This book was an early attempt
to demonstrate rational guidelines for task-oriented motor training in a format
enabling clinical testing. We were aiming to present not another approach to
intervention to compete with the eponymous approaches of the mid twentieth
century but an illustration of what would undoubtedly be a new direction for
neurophysiotherapy. This direction would involve ongoing changes to practice
based on developments in scientific understanding and the results of clinical
outcome trials. The current book represents a considerable development, since
the amount of material relevant to motor training is increasing exponentially,
and evidence of what interventions are effective and what are not is beginning
to emerge. It has become necessary for us to reconsider methods of practice for
which there is no positive evidence of effectiveness and for which there is little
scientific basis. As a result practice is making considerable change and the neuro
physiotherapy in many rehabilitation units today would be unrecognizable to
the forerunners of our profession.
There is currently, therefore, considerable focus in physiotherapy on the
need for evidence-based practice, defined by Sackett and colleagues (2000) as
'the conscientious, explicit and judicious use of current best evidence in mak
ing decisions about the care of individual patients'. As they point out, a con
certed effort is being made to find answers to questions that continually arise
in clinical practice. The best evidence is provided by randomized controlled
trials and systematic reviews. However, clinical trials and systematic reviews are
only as good as the scientific rationale underlying the intervention and the
design of the trials themselves. It is unlikely, for example, that we can develop
evidence-based physiotherapy practice if we have not first developed scientific
ally rational practice. Development of the hypothesis and its embedding within
a coherent and rational body of knowledge is a critical element in research. In
addition, for research findings to be meaningful, the detailed nature of the inter
ventions tested must be clear (methods, intensity, dosage). Where standardized
guidelines and protocols are tested, the reader knows what it is that is effective
and what is not.
As Herbert and colleagues point out (2001), basing clinical practice on evi
dence differs from traditional models of practice in which there may be prior
ity given to clinical experience as a form of evidence. With measurement of
progress becoming mandatory in the clinic, it is becoming more acceptable for
clinicians to follow a clearly identified set of guidelines as an aid to identifying
the interventions that were carried out.
Physiotherapy, together with the other health professions, is undergoing
a large degree of change with rapid growth in technology and changes in

methods of delivery. We hope this text will be an aid to clinicians and give
them food for thought.
2003 Janet Carr
Roberta Shepherd
We particularly wish to thank the neurological physiotherapy teaching team at The
University of Sydney for many stimulating discussions over the years and for their
ongoing comments on our work: Dr Louise Ada, Dr Colleen Canning, Dr Cath Dean,
Virginia Fowler and Dr Sharon Kilbreath. We particularly thank Dr Kilbreath for dis
cussions on early drafts of several chapters.
We also thank the people who so kindly agreed to be photographed for this book,
and their physiotherapists from Sydney hospitals who gave us generous support, in par
ticular Karl Schurr, Simone Dorsch and colleagues at Bankstown-Lidcombe Hospital;
Sarah Crompton and colleagues at Prince Henry Hospital; Lynne Olivetti and col
leagues at War Memorial Hospital; also Ruth Barker, University of Queensland, Claire
Lynch, Physiotherapy Department, Ipswich Hospital, Queensland, and Fiona Mackey,
Physiotherapy Department, Illawarra Area Health Service.
David Robinson provided valuable assistance and advice with photography. Dr Jack
Crosbie, Head of School of Physiotherapy, The University of Sydney, provided techni
cal assistance in the collecting of some biomechanical data.
The authors and publishers express their appreciation for being granted permission to
reproduce figures as indicated throughout the book. We also thank PRO-ED Inc., Austin,
Tx for granting permission to reproduce Figures 1.1, 2.11b, 3.17b, 3.26, 5.9b, 5.12
and 5.19c.
1 Introduction
1 Brain reorganization, the rehabilitation environment,
measuring outcomes 3
Brain reorganization,
the rehabilitation
measuring outcomes
Introduction Measurement
Measures of motor performance
Brain reorganization and
Functional outcomes
functional recovery
Circuit training: exanlple
The rehabilitation environment
Structuring a practice environment
Optimizing skill
Optimal functional recovery is the ultimate aim of neurorehabilitation after
acute brain lesion. The focus of this book is on the contribution of physiother
apy to this process using stroke as a model. The unique contribution of physio
therapy to the rehabilitation of individuals following stroke is the training of
motor control based on a contemporary understanding of impairments and
secondary adaptations, biomechanics, motor learning, exercise science and fac
tors that influence brain reorganization after injury.
The major objective of physiotherapy is the optimization of motor perform
ance in functional actions. Guidelines for task- and context-specific exercise
and training of critical everyday actions based on a scientific rationale and on
clinical research are outlined in Chapters 2-5. In these chapters, the explica
tion of normal and impaired motor performance is largely based on biomech
anical and muscle data. The guidelines emphasize the training of effective
functional motor performance using methods that provide both a stimulus to
the acquisition of skill, and increase strength, endurance and fitness.
Biomechanics provides the major foundation for understanding motor perform
ance. Without a knowledge of biomechanics, the clinician has only observation
upon which to base the analysis and training of motor performance. It is
widely recognized that analysing complex actions without a knowledge of bio
mechanical characteristics is difficult and may be inaccurate. Biomechanics
also provides methods of measuring motor performance in clinical research,
enabling the formulation and testing of interventions.
Musculoskeletal anatomy has, from the origins of physiotherapy, been a crit
ical area of knowledge for clinicians. Combining a knowledge of anatomy
and of the dynamics of multisegment movements is now a major requirement
for physiotherapists who wish to play a significant motor training role in
At this stage of the development of neurorehabilitation, the focus is on the need
for evidence-based clinical practice. Currently there is a wide variation in prac
tice and continued controversy about methods used. To some extent this may
be due to a lack of standardized treatment guidelines and protocols, together
with a paucity of outcome measurement in rehabilitation centres, which results
in little ongoing feedback about the effectiveness (or otherwise) of interventions.
This can result in a lack of awareness of the need to change or fine tune practice
to keep up with new information.
One aim in writing a text of guidelines is to illustrate an important step in the
process of improving patient outcomes: the translation of scientific knowledge
and clinical evidence into clinical practice. There is now increasing evidence
about which interventions are effective and which are not that allows clinicians
to make considerable change to methods of delivery. As Rothstein (1997) has
urged, once there is some evidence it is critical that this evidence is used in clin
ical practice.
One of the difficulties in doing clinical research and collecting data for clinical
audits is the categorization and standardization of interventions, their fre
quency, duration and intensity - what could be called the dosage. [n this book
we have categorized interventions, basing the guidelines on the available empir
ical data that are known to us. Where available these data are used to support
the guidelines; where they are not, reliance is placed on biological rationale.
This chapter starts with a brief examination of the interaction between post
lesion brain reorganization and the process of there are no
widely available medical or surgical interventions that consistently reduce the
extent of neuronal damage following stroke, rehabilitation remains the main
stay of improving outcome in these patients'> Despite its importance, until
recently very little research had been undertaken on understanding how ther
apy input may influence the process of recovery from stroke at the neurological
level. Functional brain imaging techniques now provide a way of studying
in vivo both cerebral plasticity after stroke and the influence of interventions
on the reorganization and on functional recovery.
Since there are signs that physical activity in an enriched environment facili
tates brain reorganization, this chapter also examines the typical rehabilitation
environment, suggesting ways of providing more effective and efficient meth
ods of service delivery and of building an environment that promotes physical
and mental activity and skill acquisition. In such an environment the individual
can be an active learner with opportunity for intensive exercise and training.
Collection of objective data in the clinic using valid and reliable tests is critical
for the development of effective interventions. Valid, reliable and functional
tests of motor performance are included in Chapters 2-5. This present chapter
includes a section on measuring quality of life and self-efficacy since it is now
recognized that patients' views about their functional abilities and their own
perceptions of their lifestyle are essential to planning future follow-up and
New brain imaging techniques are making it clear that the neural system is
continuously remodelled throughout life and after injury by experience and
learning in response to activity and behaviour (e.g. Jenkins et al. 1990,
Johansson 2000, Nudo et al. 2001). The possibility that neural cortical con
nections can be remodelled by our experience was suggested by Hebb over half
a century ago. Modern cortical mapping techniques in both non-human and
human subjects indicate that the functional organization of motor
cortex is much more complex than was traditionally described\Nudo and col
leagues summarize this complex organization, pointing out the extensive over
lapping of muscle representation within the motor map, with individual muscle
and joint representations re-represented within the motor map, individual cor
ticospinal neurons diverging to multiple motoneuron pools, and horizontal
fibres interconnecting distributed representations. It is their view that this com
plex organization may provide the foundation for functional plasticity in the
motor cortex (Nudd et al. 2001)
-<Support for the hypothesis that changes in the nervous system may occur
according to patterns of use comes from studies of human subjects performing
actions in which they are skilled}For example, the sensorimotor cortical repre
sentation (map) of the reading finger is expanded in blind Braille readers
(Pascual-Leone and Torres 1993) and fluctuates according to the extent of
reading activity (Pascual-Leone et al. 1995). Right-handed string players
show an increased cortical representation of flexor and extensor muscles for
the fingers of the left but not the right hand, and, with regular performance,
the representational area remains enlarged (Elbert et a1. 1995). Such studies
reflect the changes associated with skill development that are provoked by
active, repetitive training and practice, and by the continued practice of the
Conversely, restriction of activity, examined in association with immobiliza
tion (Liepert et al. 1995) or amputation (e.g. Cohen et al. 1991, Fuhr et al.
1992), also induces alterations in cortical representation. A significant decrease
in the cortical motor representation of inactive leg muscles was found after
4-6 weeks of unilateral ankle immobilization and was more pronounced when
the duration of immobilization was longer (Liepert et a1. 1995). Early investi
gation following amputation of part of one upper limb found that the remain
ing muscles in that limb received more descending connections than the
equivalent muscles of the intact limb (Hall et a1. 1990). It seems, therefore,
that the neural system is inherently flexible and adaptive, responding according
to many factors, including patterns of use.
Throughout the past century, neuroscientists have attempted to understand the
neurophysiological and neuroanatomical bases for functional recovery after
brain injury. It was assumed that other parts of the motor system must 'take
over' the function of the damaged cortex but neural models for recovery of
function were based on poorly processes (Nudo and Friel 1999).
(Over the past 10-15 years, however, neuroimaging and non-invasive stimula
tion studies - positron emission tomography (PET), functional magnetic reson
ance imaging (fMRI) and transcranial magnetic stimulation (TMS) - have
confirmed that the cerebral cortex is functionally and structurally dynamic,
that neural reorganization occurs in the human cortex after stroke, and that
altered neural activity patterns and molecular events influence this functional
reorganization (Johansson 2000))
speaking, two types of process have been suggested to underlie func
tional recovery from hemiparetic stroke: reorganization of affected motor
regions and changes in the unaffected hemisphere)following a lesion, there are
a number of physiological, pharmacological and anatomical events that occur in
widespread regions of the cortex (Schiene et al. 1996), ranging from the intact
peri-infarct zone to more remote cortical motor areas of the unaffected hemi
sphere (Furlan et al. 1996, Cramer et al. 1997). Potential mechanisms include
changes in membrane excitability, growth of new connections or unmasking of
pre-existing connections, removal of inhibition and activity-dependent synaptic
changes. Plastic changes have also been demonstrated in subcortical regions.
Knowledge of the potential capability of the brain to adapt following a lesion and
of the probability that patterns of use and disuse affect this reorganization pro
vides a stimulus to the design of optimal stroke rehabilitation (Carr and Shepherd
1998, Shepherd of animals and humans with brain lesions provide
insight into the process of functional recovery and reflect a relationship between
neural reorganization and the rehabilitation process. A series of studies of squirrel
monkeys that modelled the effects of a focal ischaemic infarct within the hand
motor area of the cortex, found that there was further loss of hand representation
in the area adjacent to the lesion when monkeys had no post-infarct intervention
(Nuda and Milliken 1996). This suggested that further tissue loss could have
been due to non-use of the hand and this was confirmed by a follow-up study
which showed that not only could tissue loss be prevented when the unimpaired
hand was restrained and the monkey had daily repetitive training in skilled use of
the impaired hand, but there was also a net gain of approximately 10% in the
total hand area adjacent to the lesion (Nuda et al. 1996))In a further study, in
which the unimpaired hand was restrained but no training was given, the size of
the total hand and wrist/forearm representation decreased, indicating that
restraint of the unimpaired hand alone was not sufficient to retain the spared
hand area (Friel and Nudo 1998))The results of these studies point to the neces
sity for active use of the limb for the survival of undamaged neurons adjacent to
those damaged by cortical injury, and suggest that retention of the spared hand
area and recovery of function after cortical injury might depend upon repetitive
training and skilled use of the hand (Nudo and Friel 1999).
Human studies also provide evidence for functional plasticity after stroke asso
ciated with meaningful use of a limb (for review see Cramer and Bastings 2000)
~ J
: of
, In
: of
involving task-oriented repetitive exercise (Nelles et al. 2001). The results of a
recent study using TMS suggest a relationship between physiotherapy interven
tion and reorganization of the cerebral cortex in persons several years after
stroke (Liepert et al. 2000). The authors examined the output area of the
abductor pollicis brevis muscle in the affected hemisphere after constraint
induced (CI) therapy which involved restraint of the unimpaired hand and
training of the impaired limb. They reported asignificant enlargement in the
size of this area which was associated with improved motor performance of the
impaired hand. These changes were maintained 6 months later, with the area of
the cortical representation in the affected hemisphere almost identical to the
unaffected side. The results of several other studies also support the idea that
there is considerable scope for use-dependent functional reorganization in the
adult cerebral cortex after stroke (e.g. Weiller et a1. 1992, Cramer et a1. 1997,
Kopp et al. 1999).
It has been noted that soon after stroke the excitability of the motor cortex is
decreased and cortical representations ate reduced. This probably occurs as a
result of diaschisis-like effects (Nudo et al. 2001). The recovery of function
occurring early following stroke reflects reparative processes in the peri-infarct
zone adjacent to the injury.<fhese include the resolution of local factors such as
oedema, absorption of necrotic tissue debris and the opening of collateral chan
nels for circulation to the lesioned area. It is thought that this takes place over a
relatively short period, perhaps 3-4 weeks (Lee and van Donkelaar 1995).
However, the principal process responsible for functional recovery beyond the
immediate reparative stage is likely to be use-dependent reorganization of
neural mechanisms. What is certain is that adaptive plasticity is inevitable after
an acute brain lesion, and it is very likely that rehabilitation may influence it,
with the possibility of positive or negative consequenceS)
It is reasonable to hypothesize that repetitive exercise and training in real-life
tasks following a stroke may be a critical stimulus to the making of new or
more effective functional connections within remaining brain tissue. Training
and practice using methods that facilitate motor learning or relearning would be
essential to the formation of new functional connections. Increase of synaptic
efficacy in existing neural circuits in the form of long-term potentiation and for
mation of new synapses may be involved in earlier stages of motor learning
(Asanuma and Keller 1991). It seems certain from current research that if rehabili
tation is to be effective in optimizing functional recovery, increased emphasis
needs to be placed on methods of forcing use of the affected limbs through
functionally relevant exercise and training, including specific interventions such
as treadmill walking and constraint-induced training of the affected upper limb.
There are many complex issues arising from the studies cited here and from
others. However, the evidence to date is very compelling in terms of the import
ance of intensive training in behaviourally relevant tasks in a stimulating and
enriched environment after cortical injury. \What is certain is that the brain
reorganizes after injury. The changes, however, may be either adaptive or mal
adaptive in terms of functiOlyGiven the evidence from differential and context
dependent effects of reorganization, it is possible to hypothesize that the nature
of organization depends on the inputs received and outputs demanded. To what
extent functional imaging data correspond to outcome data needs further eval
uation (Nelles et al. 1999). Ifwe can understand the underlying processes and
the interactions between brain reorganization, training and motor learning,
interventions can be designed on neurobiological principles that may direct
intact tissue to 'take over' the damaged function.
Although there is mounting evidence indicating that patterns of use affect
reorganization, and increasing acceptance of a link between brain reorganization
and recovery of function, it is still not generally accepted in medical or therapy
practice that there might be a link between specific rehabilitation methods and
functional recovery. Many clinicians would argue that rehabilitation is critical
after stroke but only a few would take the next step and argue that it is the
nature and the process of rehabilitation and the methods used that affect recov
ery, that some methods in use may be a waste of time or may actually inhibit
In summary, there is evidence of, substantial potential for functional brain
reorganization after stroke. Although the exact mechanisms are still unclear,
results from animal and human studies suggest that a substantial substrate
exists for influencing the recovery of an injured brain. Although the size of tis
sue loss has been equated with outcome, Johansson (2000) comments that 'it is
not only the number of neurons left, but how they function and what connec
tions they can make that will decide functional outcome'.
If brain reorganization and functional recovery from brain lesions are depend
ent on use and activity, then the environment in which rehabilitation is carried
out is likely to play an important role in patient outcomes. The rehabilitation
environment is made up of:
1. the physical or built environment (the physical setting)
2. the methods used to deliver rehabilitation
3. the staff, their knowledge, skill and attitudes.
We argue in this section that the physical environment must offer possibilities
for intensive and meaningful exercise and training; that the delivery methods
should vary and include supervised one-to-one practice with the therapist,
group practice and independent practice; and that therapists can be major
facilitators of motor learning.
Evidence from animal experiments suggests that the nature of the environment
and its physical structure, together with the opportunities it offers for social
interaction and physical activity, can influence outcome after a lesion. Environ
mental enrichment after a brain lesion in rats can increase cortical thickness,
protein content, dendritic branching, number of dendritic spines and size of
synaptic contact areas (Kolb and Gibb 1991, Pritzel and Huston 1991, Will and
Kelche 1992). Conversely, the mechanisms of brain plasticity are also affected
by an impoverished environment. It has been suggested that environmental
enrichment may affect a general factor or a host of specific factors that relate to
the general adaptive capacity of the organism, i.e. its ability to cope with a var
iety of situations and problems of a general nature (Finger 1978).
Animal experiments have shown that exposure to a particular type of environ
ment may either enhance or restrict opportunity for general experience. Rats
housed in an enriched environment, with the opportunity for physical activities
and interaction with other rats, performed significantly better on motor tasks
than rats housed in a standard laboratory environment, alone and with no oppor
tunities for interesting activities (Held et al. 1985, Ohlsson and Johansson 1995).
In animal research, the aspects of the enriched environment which appear to be
critical as enhancers of behaviour are social stimulation, interaction with
objects that enable physical activity (Bennett 1976) and an increased level of
arousal (Walsh and Cummins 1975). It is of interest, therefore, to consider
what effects a typical rehabilitation environment may have on the behaviour of
humans, and observational studies of rehabilitation settings provide some
insights. They suggest the environment may not be sufficiently geared to facili
tating physical and mental activity or social interaction, and that the rehabili
tation setting may not function as a learning environment (Ada et al. 1999).
Several studies investigating how stroke patients spend their time have shown
that a large percentage of the day was spent in passive pursuits rather than in
physical activity. Even less physical activity occurred on weekends compared
with weekdays and there was little evidence of independent exercise. Patients
were noted to remain solitary and inactive for long periods, watching others or
looking out of the window. Although what physical activity occurred was in
the therapy area, therapy was noted to occur for only a small percentage of the
day (Keith 1980, Keith and Cowell 1987, Lincoln et al. 1989, Tinson 1989,
Mackey et al. 1996, Esmonde et al. 1997). It is disappointing to see that over
the nearly two decades spanned by these studies there was little change in the
amount of time spent in physical activities.
Structuring a practice environment
The goals of physiotherapy are to provide opportunities for an individual to
regain optimal skilled performance of functional actions and to increase levels
of strength, endurance and physical fitness. For the able-bodied and the dis
abled, it is recognized that practice is the way to achieve these aims.
Skill in performing a motor task increases as a function of the amount and
type of practice as does strength, endurance and fitness. Although the amount
of practice is critical and more practice is better than less, recent research into
motor learning and strength training suggests that the amount of practice may
not be the only variable influencing any improvements in performance. There
is some clinical evidence (e.g. Parry et al. 1999a) that merely increasing the
time spent in therapy may not necessarily improve outcome. It is probably the
type of practice rather than simply its occurrence that reshapes the cortex fol
lowing a brain lesion (Small and Solodkin 1998).
The issue of how much time is spent in physical activity, including practice of
motor tasks and how this time is organized, is a critical one for rehabilitation.
The results of observational studies on how patients spend their day raise the
question of whether the typical rehabilitation environment is sufficiently stimu
lating, providing opportunity for intensive and meaningful training and exer
cise. In other words, are the physical structures, organization and attitudes
required to encourage active participation, exercise, independent practice and
learning, in place? If they are not, a change in both staff attitude and work
practices may need to be considered.
What the patient is actually doing in rehabilitation must itself be effective if
increasing the amount of time spent practising is to improve outcome and there
is increasing evidence of which methods can be effective. What the patient prac
tises outside the time allotted to training and practice is also likely to impact sig
nificantly on outcome. As an example, self-propulsion in a one-arm drive
wheelchair using unimpaired limbs is at odds with treatment goals to increase
strength and control of impaired limbs (Esmonde et al. 1997). If the patient
spends more time in this activity than in exercising the impaired limbs, it is not
hard to guess the probable outcomes.
Delivery of physiotherapy
It is clear from studies of rehabilitation environments that only a small part of
the day may be made available for practice and training. To some extent this
reflects work practices. One way of increasing the amount of training and
practice is for therapists to move away from reliance on hands-on, one-to-one
therapy, to a model in which the patient practises, not only in individualized
training sessions with the therapist, but also in a group with less direct therapist
supervision (Fig. 1.1). One or two therapists with the assistance of an aide can
oversee several individuals, with relatives and friends helping if they are will
ing. Such organization can improve efficiency of delivery by increasing the time
the patient spends practising without substantially increasing the therapist's
load. This ensures better use of the short time allowed for intensive inpatient
rehabilitation (Fig. 1.2).
Group training can enable each individual to exercise independently or with
another patient at specially designed work stations set up to enable practice of
specific motor activities and with technological and other aids to promote
strength and skill. Work stations can provide videotapes and/or photographs
of the action to be practised, lists of some common problems and what to do
about them.
Group training and exercise can be organized as a circuit class (see note at end
of chapter). Circuit training in able-bodied individu.als often involves two people
working together, one observing and one physically practising. Alternating
practice with a partner can be more effective than individual practice (Shea
et al. 1999). Practising with another person in an interactive way may increase the
learner's motivation by adding both competitive and cooperative components to
the practice situation (McNevin et al. 2000), and there are benefits to be gained
from observing another person learning a task (Herbert and Landin 1994).
Brain reorganization 11
FIGURE 1.1 Group training: (a) Step-up exercise class. Safe independent practice is ensured by
having a high table nearby for support when necessary. (b) Standing up and sitting
down class. Seat height is customized to individual's lower limb strength.
ed (b)
12 Introduction
FIGURE 1.2 Indirect supervision. (a) Simple reaching tasks in a harness can be progressed to
reaching in a tandem foot position and to stepping out in order to reach further.
(b) Repetitive practice of standing up and sitting down. (c) Repetitive leg exercises
on a pedal machine. The paretic foot can be bandaged on to the pedal if necessary.
(a) (b)
Brain reorganization 13
FIGURE 1.3 Overview of an upper limb circuit class at a Queensland hospital. The work stations
include reaching to attach stick-ons high on the wall, pegging playing cards on a line,
modified (supported) reaching to push bean bag over the edge, playing a game with
an opponent. (Courtesy of C Lynch, Physiotherapy Department, Ipswich Hospital.)
Working with another person provides rest intervals, especially when training
is physically demanding, yet the individual has to remain focused in order to
assist his/her partner. Moving on from one set of tasks to another in a circuit
class is also a way to keep up alertness and enthusiasm.
These methods of delivery fit well within an exercise and traInIng model of
rehabilitation, and in the clinic such group training is being used by therapists
(Fig. 1.3), with evidence that exercise classes and circuit training can be both effi
cacious and feasible (Taub et a!. 1993, Dean et a!. 2000, Teixeira-Salmela et al.
1999, 2001). Although the therapist's supervision is necessary, patients may
benefit also from the experience of their peers (McNevin et al. 2000) and from
helping each other. The presence and input from other patients with similar
disabilities may motivate and encourage individuals to do better. In a recent
study, it was found that the introduction of group exercise classes increased the
level of social interaction throughout the day (De Weerdt et al. 2001). Such par
tially supervised practice can bridge the gap between individualized training and
unsupervised practice in hospital and at home. Giving patients some control over
practice conditions by empowering them to view the therapist as a resource per
son may encourage their problem-solving abilities (McNevin et al. 2000).
Actively involving patients and giving them a sense of responsibility in their
exercise and training programme enables them to take some control over
events that affect their lives. Although the patient's level of disability would
influence the amount of independent practice that is possible, once the ther
apist and patient have decided how much practice can be expected, strategies
to enable this to occur can be developed. Brief exercise lists with diagrams
for independent practice, stored on computer, can be printed as appropriate
for the individual. Monitoring needs to be put in place to determine whether
or not the desired activity levels are being achieved, and individual work books
with simple check lists can provide evidence of what each person has done and
for how long in 1 day, or 1 week. There have been reports that, from the
patients' point of view, perceived barriers to exercise programmes include lack
of positive feedback and the perception that exercises are not adapted to their
particular situation or daily routine. Patients who reported that their physio
therapists were very satisfied with progress were more likely to participate than
patients who did not know whether or not the therapist valued their efforts
(Sluijs et al. 1993).
Some therapists express concern about reducing hands-on therapy and increas
ing semi-supervised and unsupervised practice because patients may practise
'incorrectly'. However, it is well known that people do not learn to perform an
action unless they have a chance to practise it themselves. It is only by attempt
ing to carry out a movement to achieve a goal voluntarily that an individual
knows whether the action attempted is successful. Motivation is known to be
low when the learner cannot make mistakes during practice. In other words,
learning requires an element of trial and error. Practice conditions can be orga
nized to enable practice of the action even in patients with muscle weakness
and poor limb control. For example, walking on a treadmill with partial body
weight support has been shown to be more effective in improving gait than
restricting walking practice in order to avoid the patient practising incorrectly,
i.e. making errors (Hesse et al. 1995).
There is increasing understanding of the importance of strength training and
exercise in stroke rehabilitation. However, the intensive and repetitive practice
required to improve performance may not be demanded if patients are perceived
to fatigue easily, particularly if they are elderly. Fatigue may be viewed as con
traindicative and necessitating rest, instead of as a manifestation of the need to
get stronger and develop more endurance. The past decade has seen an accumu
lation of data that provide reassurance of the safety of exercise and physical
conditioning programmes following stroke (Ch. 7). The beneficial effects of such
programmes include not only improved physiological responses but also
improved functional motor performance. Simple ways to increase exercise toler
ance and endurance, even in the early stages, may include setting goals such as
increasing the speed of movement and the number of repetitions. The results can
be graphed, providing the patient with feedback information and motivation.
Challenges to customary delivery of intervention can be difficult for clinicians
because such practices have been built on long-standing traditions and beliefs
and a culture which has not until recently demanded evidence of efficacy. The
attitude that the critical method of therapy delivery is a one-to-one interaction,
with emphasis on physical handling of the patient in order to facilitate more
normal movement, does not fit well with modern views of the patient as active
learner and the therapist as teacher or coach. Increasing numbers of clinical
studies are supporting the use of more varied methods of delivery and more
active methods of intervention.
Optimizing skill
Skill has been defined as 'any activity that has become better organized and
more effective as a result of practice' (Annett 1971) and also as 'the ability
to consistently attain a goal with some economy of effort' (Gentile 1987).
Everyday activities, even such apparently simple ones as standing up from a
seat, constitute motor skills. They are complex actions, made up of segmental
movements linked together in the appropriate spatial and temporal sequence.
The person with motor impairments following a neural lesion needs to learn
again how to control segmental movement so that the spatial configuration
and temporal sequencing of body movements brings about an effective action,
thereby achieving the individual's goal with minimum energy expenditure.
Task-oriented training is critical to gaining the necessary control. In attempting
to regain movement control, however, the degree of muscle weakness may be
such that initially the patient also has to practise exercises for activating and
sustaining activity in specific muscle groups.
An important characteristic of acquiring skill is that learning seems to take
place in overlapping stages. Two models that have been proposed give insights
useful to the clinic. An early study described three stages of learning: an early or
cognitive stage, an intermediate or associative stage and a final or autonomous
stage (Fitts and Posner 1967). The learning stages have also been described as
first getting the idea of the movement, then developing the ability to adapt the
movement pattern to the demands of the environment (Gentile 1987). In both
cases the initial stage is cognitive.
Early after stroke, patients with a lesioned system are struggling to learn again
how to perform even simple movements as well as everyday actions. They need
to practise repetitively in order to get the idea of the action they are (re)learn
ing and to train the neural coordination necessary for effective performance.
As they gain more strength and control, less attention can be directed toward
performing the action and more attention to the goal and relevant environmen
tal cues.
As motor control and skill develop, changes occur at different levels. For
example, with more intersegmental control there is less tendency to 'freeze' the
degrees of freedom. There is also a decrease in energy expenditure. The focus of
attention shifts. In walking, the focus of visual attention shifts from the feet to
the surrounding environment; 'star billing' for sit-to-stand changes from foot
placement and speed of trunk rotation to steadying a glass of water while stand
ing up. Being aware of the characteristics of each stage enables the therapist to
provide the appropriate practice conditions to optimize performance.
For movement to become more automatic, intensive practice is required in
order to increase endurance as well as the opportunity to practise adapting the
action to varying environmental demands - for example, walking in different
environments, coping with several simultaneous demands such as doing two
things at once.
Focusing attention
Attentional abilities required for everyday functioning include the ability to
focus attention on a task, to sustain concentration, to shift attention if add
itional task-relevant material comes along but to ignore distracting, irrelevant
inputs. Training techniques to overcome these deficits involve highlighting the
task to bring it to the patient's attention, reinforcing maintenance of attention
to task and its relevant cues, and adding distractions to the environment then
reinforcing the 'filtering out' of such inputs (Adams 1990).
Learning of motor skills involves two critical components, particularly in the
early stages:
1. identifying what is to be learned
2. understanding the ways through which the goal can be accomplished.
A number of recent studies demonstrate that the person's attentional focus can
have a decisive influence on motor performance and learning (Wulf et al.
1999a). It is important to decide what the patient should pay attention to dur
ing practice. Two methods of directing the focus of attention are demonstra
tion (both live and recorded) and verbal instruction.
Demonstration. Demonstration, or modelling of an action, gives the therapist the
opportunity to point out the key components of the action. A demonstration
assists the patient to get the idea of the spatial and temporal features of the
action, the 'shape' or topology of the movement (Fig. 1.4).
Instructions. In the early stages of learning, instructions should be brief, with no
unnecessary words. They should cue the patient to concentrate on one or two
critical features of the action (Fig. 1.5). In standing up practice, for example,
the patient is drilled to place feet back, swing upper body forward at the hips
and stand up, and can verbalize this sequence while practising. A stick figure
drawing can reinforce these key elements. Asking the patient to list what cues
they are to focus on aids mental rehearsal and enables the therapist to judge
what the patient has gained from both instruction and demonstration.
Gradually the patient internalizes these details and can start to think more of
the goal of the action (e.g. to stand up and walk to another chair) while prac
tising in a more complex environment.
At this later stage, instructing the patient to focus on the details of their perform
ance may actually be detrimental. There is increasing evidence that directing
learners to the effects of their movements, i.e. whether or not the goal is
achieved, may be more effective than attending to the movement itself (Singer
et al. 1993, Wulf et al. 1999a). Similarly, at this stage of learning, verbal
instructions are used in directing attention to specific environmental features or
Brain reorganization 17
FIGURE 1.4 Demonstrating the extent of trunk f'exion at the hips in standing up: (a) beside the
patient so that he can line his shoulders up with hers; (b) a sagittal plane view to show
him how far forward the shoulders are moved.
(a) (b)
FIGURE 1.5 The drawing illustrates in a simplified form the critical features of the action.

A typical approach to the question of what a person should attend to during
practice has been to tell learners to be aware of what they are doing by paying
attention to the 'feel' of the movement (Singer et a!. 1993). This has also been a
common approach in clinical practice. Singer and colleagues (1993) compared
the influence on performing and learning a throwing task of consciously attend
ing to the movement (awareness strategy) or focusing on one situational cue
and ignoring the movement (non-awareness strategy). The findings indicated
that, for able-bodied subjects, the non-awareness strategy led to higher achieve
ment than the awareness strategy. Similar results have been found in several
recent studies (see Wulf et al. 1999b for discussion). It may, therefore, be bene
ficial at times for the therapist to suggest a focus of only one cue, following this
with a 'Now - just do it'.
Setting goals. The goal to be achieved should be meaningful and worthwhile to
the individual, reasonably challenging yet attainable. Some goals seem to facili
tate action more readily than others. As an example, tasks that have goals
directed toward controlling one's physical interaction with objects or persons in
the immediate environment (concrete tasks) seem to have more meaning and be
more motivating than goals directed at movements for their own sake (abstract
tasks). Performance has been shown to improve significantly when an individual
is presented with a concrete goal compared with an abstract goal. The facilita
tory effect of a concrete goal has been demonstrated in several clinical studies,
including one study of adults with restricted range of motion in the elbow or
shoulder as a result of injury (Leont'ev and Zaporozhets 1960), and another of
children with cerebral palsy (van der Weel et al. 1991). Furthermore, kinematic
analysis has clearly demonstrated the different movement characteristics when a
task is performed without the relevant object (Mathiowetz and Wade 1995, Wu
et al. 2000). Wu et al. (2000) examined a group of individuals after stroke and
an able-bodied group scooping coins off a table with and without the coins pres
ent. For both groups, reaching for the coins elicited faster movement compared
with reaching without the coins. The data showed a greater percentage of reaches
where peak velocity occurred, and smoother and straighter movement, all signs
of a coordinated well-learned movement.
There may be several explanations for these results. Abstract and concrete tasks
differ in the degree to which the action is directed toward controlling physical
interaction with an object as opposed to merely moving a limb. The presence of
an object provides clear visual information before starting the movement as well
as providing feedback about performance and outcome. A functional task that
is also familiar is more likely to produce more effective performance than an
abstract task. A task that is relatively difficult to attain but achievable may pro
vide more motivation, challenging the patient to succeed.
It has been rather a common practice in physiotherapy to have the patient
practise abstract movements such as weight shift sideways in standing rather
than the more concrete and meaningful task of reaching sideways for an
object, or placing one foot on a step. In the former, the patient has to 'believe'
the therapist's feedback about whether or not the goal was achieved. For the
latter, there are observable outcomes. Examples of providing concrete goals to
improve functional performance are given in Chapters 2-5.
An essential aspect of skill acquisition is the feedback that learners receive
about their performance of an action. There are two principal forms of
feedback: intrinsic, which is the naturally available sensory feedback (visual, pro
prioceptive, tactile) occurring as part of the action, and augmented feedback,
providing knowledge of the results of the action (KR) and knowledge of the per
formance itself (KP). Augmented feedback comes from external sources, such as
the therapist or coach, and adds to or enhances intrinsic feedback. Augmented
feedback can be positive or negative, qualitative or quantitative, and can provide
motivation when presented in a way that influences the learner to persist in striv
ing toward the goal.
The most commonly used form of augmented feedback is verbal feedback from
the therapist or teacher. Feedback can involve instrumentation, and may
include videotape replay or graphed representation of kinematics (path of body
parts, joint angle displacement) or kinetics (amplitude and timing of vertical
force) as a means of providing information about some aspect of performance.
Electromyography (EMG) biofeedback is a commonly used method of provid
ing information about physiological processes underlying the action.
Augmented feedback can have various effects on skill performance depending on
the type of information given, how it is given and the skill being learned. In pro
viding feedback about performance, the challenge for the therapist is to select the
appropriate feature of performance on which to base that feedback. It is, there
fore, important to prioritize biomechanical components so that feedback is con
fined to these key characteristics of the movement. The patient's attention is
directed to the part of the action to be changed in order to improve the next
attempt. As skill in performance increases, feedback changes from prescriptive in
the early stages to descriptive, which may be more appropriate in the later stages.
The experience a person has in making and correcting errors is part of skill learn
ing. Not every attempt needs to be 'perfect'. There is a continuing controversy
about the role of augmented feedback in terms of whether it should emphasize
errors made or aspects of performance that are correct. Magill (1998) writes that
augmented feedback (as error information) functions in an informational role
that is related to facilitating skill improvement. Augmented feedback about what
the person did correctly has a more motivating role but may not be sufficient to
produce optimal learning, particularly in the early phase (Magill 2001).
If practice conditions create a dependency on augmented feedback, the individ
ual may rely on it with the result that performance deteriorates and learning is
impeded when it is removed (Magill 2001). Individuals may learn to depend
on the therapist or the instrumentation for guidance instead of attending to
task-relevant cues (Sidaway et al. 2001). There is evidence in clinical studies of
this reliance on instrumented feedback and subsequent deterioration of perform
ance when feedback is removed (Engardt 1994, Fowler and Carr 1996). The
equivocal results of many clinical studies on instrumented feedback may be largely
due to this problem of withdrawal. Overreliance on feedback may be avoided if
the patient is assisted to replace augmented feedback by attending more to the
action itself, i.e. was the goal achieved easily and quickly? If not, then what
was the reason?
Reducing the number of trials for which feedback is given is likely to result
in better learning than feedback after every trial. There is some evidence that
providing feedback after each attempt may not be as useful as encouraging the
patient to ask for feedback whenever they think it might be beneficial (Janelle
et al. 1997). The active role of the learner has been neglected in the study of feed
back until recently. Giving the learner some responsibility in and control over the
learning situation may encourage the individual to engage in more effective
learning strategies and increase confidence and motivation (Janelle et al. 1997).
In a recent study, Talvitie (2000) videotaped and systematically observed physio
therapist-patient interactions during treatment and noted the way in which
feedback was given. The author reported that therapists did most of the talk
ing, patients were rarely involved in the dialogue, information feedback
appeared very seldom, and verbal and physical guidance were given more than
visual cues. The methodology of this study provides a useful way for therapists
to get feedback about their own interactions with patients. Therapeutic rela
tionships may be based on the physiotherapist as authority figure or on a more
equal footing with the patient. The latter based on dialogue helps to create the
opportunity for more positive relationships (Lundvik et al. 1999).
Vision plays a key role in motor learning and is probably the most widely used
source of information feedback in performing motor tasks. Vision provides
powerful intrinsic feedback, information about environmental conditions, and
exproprioceptive information for determining the individual's relative position
within the environment (Lee and Aronson 1974).
Early after stroke, patients typically have difficulty directing their visual atten
tion to environmental cues and the therapist may need to direct vision to the
appropriate cue. When reaching for an object, a reminder is given to look at
the object; when walking in a busy corridor, vision is directed well ahead, not
at feet or floor. As skill progresses, the individual learns to direct their own
vision and to pay attention to the critical features in the environment which
provide the most useful information for achieving the goal, and enable predic
tion of potential hazards in order to take appropriate avoidance action.
Transfer of learning
Early rehabilitation typically takes place almost entirely within the rehabilita
tion setting itself. The purpose of both strength training and task practice in
rehabilitation is, however, to increase the patient's ability to perform the
action in situations where it is needed, in both everyday and novel contexts.
A major goal of the therapist as facilitator and teacher is, therefore, to assist the
patient to transfer training (learning) from the practice environment (the rehabili
tation setting) to these other environments.
Transfer of learning from one performance situation to another is an integral
part of skill learning (Magill 2001). This concept lies at the heart of understand
ing motor learning, acknowledging that an action cannot be separated from the
environment in which it is carried out. How we move is the sum of the task
undertaken and the environment in which it is performed. Gentile's (1987) tax
onomy of motor skills is a framework for encouraging transfer of a particular
motor task into different environments. The taxonomy differentiates motor
skills depending on the environmental context. A closed motor skill is one in
which the environment is stable. If the environment is changeable, skills are cat
egorized as open. Regulatory conditions during performance are either station
ary (walking up a flight of stairs) or in motion (walking through automatic
doors); intertrial variability is either absent (walking in an uncluttered room) or
present (catching a ball). Constraints from the environment alter the bio
mechanics of an action and the amount of information to be processed. When
performance depends on external timing, the demand on information process
ing increases because success depends on predicting future events. For example,
consider the demands of crossing a road, even at traffic lights. Performance also
varies according to spatial relationships. Walking differs depending on whether
one is walking on a flat surface, a slippery surface or approaching a kerb.
A patient may learn to walk in the closed environment of the physiotherapy
area but not be able to walk outside it unless given the opportunity to practise
in open and more complex situations. Categorization of motor skills provides
the therapist with a basis on which to set up practice conditions that provide a
variety of relevant situations and challenges for training that prepare the indi
vidual for the real world. When practice is varied by changing aspects of the
environmental context, the motor skill that develops is flexible and generative,
facilitating a kind of motor problem-solving ability. Huxham et al. (2001) pro
vide some clear examples of assessing and training balance in relation to func
tion and the physical environment using Gentile's taxonomy.
Also of interest is the issue of transfer from strength training to the perform
ance of a specific action. Some years ago, Thorndike suggested that transfer
was the result of identical elements being transferred from one task to another.
Since then the motor learning and exercise science literature on able-bodied
people has provided evidence that a transfer can occur to tasks with similar
dynamic characteristics (Oxendine 1984, Gottlieb et al. 1988). Clinical litera
ture suggests that this also occurs in disabled populations (Nugent et al. 1994,
Sherrington and Lord 1997). This issue is discussed in Chapter 7.
From the literature on motor learning it is clear that, as a general rule, the
action to be acquired should be practised in its entirety (task-specific practice),
since one component of the action is to a large extent dependent on preceding
components. Several studies of gait have demonstrated that power generation
in the ankle plantarflexors for push-off at the end of stance phase is critical to
the subsequent swing phase in terms of step length. Power generation in the
ankle plantarflexors, as part of the gait cycle, also affects walking speed and
can ensure an effective pattern of energy exchange (Winter 1983, Olney et al.
1986). These mechanisms can only be optimized through the practice of walk
ing itself. The strength training literature is also clear that, above a certain
activity-specific threshold of strength, exercises should resemble as closely as
possible the action being trained.
There are many reasons why functional improvement in performance of an
action depends on practice of the action itself. Nevertheless, in rehabilitation it
22 Introduction
may be necessary for the patient to spend time in part practice, i.e. practising a
sub-task. An important question is determining which component part of the
action to practise separately. If a patient cannot activate plantarflexor muscles
with the necessary force and at the appropriate time for push-off, repetitive,
resisted exercises are necessary to strengthen (and actively stretch) these muscles.
Practice of this sub-task is then incorporated into walking during training.
When an individual has difficulty performing an action due to muscle weakness,
and can only practise that action using grossly ineffective compensations, modi
fying the task or environment is a way of reducing task difficulty. Such modifi
cations enable practice of a simplified form of the action. Raising seat height
decreases lower limb muscle force demands and may enable a person to practise
standing up and sitting down with some weight through the affected lower limb
or more evenly distributed between the two lower limbs; walking on a treadmill
with partial body weight support enables early walking practice even in a very
weak patient provided assistance is given to the swing phase; constraining
mechanically inappropriate movement with a splint or strapping enables practice
of limb loading exercises without loss of control into hyperextension (Fig. 1.6).
FIGURE 1.6 (a) A tendency for the knee to collapse into hyperextension is prevented by strapping
across the back of the joint. (b) This enables the limb to be loaded with thigh and
shank segments in an alignment which may encourage activation of lower limb
extensor muscles.
(a) (b)
In both strength trammg and skill development, repetltzon is an important
aspect of practice. Repetitive exercise and practice of an action facilitates the
contraction of the muscles involved and improves performance in both able
bodied and disabled individuals (Asanuma and Keller 1991, Butefisch et al.
1995, Dean and Shepherd 1997). Thousands of repetitions of an action may be
necessary for the patient to develop an optimal way of performing the action
(Bernstein in Latash and Latash 1994). Traditional physiotherapy neglects the
repetitive element of skill acquisition that probably forms an essential prerequis
ite in motor rehabilitation (Butefisch et al. 1995).
In the early stages of rehabilitation, repetitive practice of an exercise or part of
an action is necessary to increase muscle strength and train coordination of the
muscular synergies that move the segmental linkage. The therapist should sus
tain the patient's motivation during repetitive practice, for example, by count
ing repetitions (or providing a counter), and by providing concrete feedback of
the effects of practice in terms of increases in strength or in speed. It may be
useful to explain to the patient that learning continues even though perform
ance may deteriorate due to muscle fatigue. After a rest period or a change to a
less demanding or different activity, an improvement in performance is usually
Acquiring skill does not only mean to repeat and consolidate but also to invent
and to progress (Whiting 1980), i.e. a particular type of 'repetition without
repetition' (Bernstein 1967). Thus skill acquisition involves the ability to solve
motor problems posed by the environment. Patients need the opportunity to
practise motor actions in different task and environmental contexts in order to
develop this flexibility.
Practice can be considered as a continuum from overt physical practice at one
end of the spectrum to covert or mental practice at the other. Mental practice
is a technique that uses imagery to mentally rehearse a physical action without
overt movement of the body or limbs. A meta-analysis of research into the
effect of mental practice in able-bodied subjects found that, in general, mental
practice is more effective than no practice, although the effects reported have
been inconsistent (Feltz et al. 1988). Mental practice is likely to be most effect
ive once the individual has developed a basic understanding of the movement's
topology and when combined with physical practice. There is some empirical
evidence in healthy subjects that mental practice activates the appropriate
muscles and neural networks and helps establish and reinforce appropriate
coordination patterns (Magill 2001).
Mental practice in rehabilitation may only be suitable for patients who do not
have communication or perceptual problems and who understand the idea.
Time should be taken to ensure that the patient understands the details of the
movement to be rehearsed mentally. Observing a demonstration of the action
can help the person develop a mental picture of the action to be performed.
The patient can repeat back what he/she is imaging and have a photograph or
drawing of the action for reference. A list and/or audiotape of the steps to
remember should also be provided. Mental practice may have a role to play
in motor training when the patient has very little muscle activity. It may also
provide a means of maximizing the time spent practising outside structured
exercise sessions (Van Leeuwen and Inglis 1998, Page et al. 2001).
In summary, given the evidence that experience and use-dependent tramlOg
drive brain reorganization either positively or negatively, a rehabilitation unit
should provide:
a stimulating and challenging environment to facilitate social interaction
and active participation
facilities for intensive task- and context-specific exercise and training to
optimize motor performance and skill
a varied delivery of intervention, i.e. in groups and in circuit training,
using exercise equipment and electronic aids.
The reader is referred to reviews of motor learning theory and research for
more detail (e.g. Gentile 2000, Magill 2001).
Rehabilitation of patients after stroke remains a major challenge to the health
care system. The cost-benefit relationship of stroke rehabilitation as much or
more than humanitarian concerns has increased the need for the health-care
community to be accountable for the services they provide (Keith 1995).
There is a continuing discussion about the value of focused stroke rehabilitation
and no clear indication that patients in these programmes do better than those in
general programmes. All those involved in this controversy agree that there is not
only a need for a more scientific foundation of therapeutic intervention but also
for better documentation of the benefits of stroke rehabilitation (Mauritz 1990).
Major problems in clinical rehabilitation research are the lack of consistency in
measures used and the amount of heterogeneity in both patient groups and
interventions. It can be difficult to make useful comparisons of data from dif
ferent studies, centres and countries. A consensus on a few well-grounded out
come measures would help in examining the value of stroke rehabilitation and
texts already exist that identify valid and reliable measures. The usefulness of
rehabilitation research depends also on the inclusion of details of interventions
given and patient population in order to make a link between the methods
used in rehabilitation and the functional outcomes.
Evidence-based practice (for discussion see Herbert et al. 2001) is an important
strategy for 'meeting the ethical imperative of providing the best possible care
for our patients' (Sherrington et al. 2001). Qualitative and quantitative
research, in which measurement plays a critical part, makes a major contribu
tion to the development of evidence-based clinical practice. The best evidence
of the effects of intervention is provided by randomized controlled trials and
systematic reviews
. The Physiotherapy Evidence Database" is an internet-based
twww.cochranelibrary.net; *ptwww.fhs.usyd.edu.au/pedro
database of all randomized controlled trials and systematic reviews relevant to
physiotherapy. We agree with Sherrington and colleagues (2001) who point
out that it is of vital importance that physiotherapy clinicians ensure that evi
dence is translated into practice. A further responsibility for clinicians is the
collection of objective data. Since physiotherapy is concerned principally with
training the individual to improve performance of motor actions, motor per
formance is the focus of measurement.
Measures of motor performance
It is essential in everyday clinical practice that there is a systematic examin
ation of the effects of interventions in order to document patient progress and
improve outcomes. Anecdotal evidence suggests that many physiotherapists do
not systematically evaluate their treatment methods and therefore do not
obtain any objective feedback about their work.
Interesting insights come from a recent follow-up survey of Canadian physio
therapists (Kay et al. 2001). Following a deliberate focus in continuing educa
tion on methods of measurement, therapists surveyed were more aware of the
complexities of outcome measurement than they were in the previous survey
(1992). However, they reported that they lacked confidence and the skills neces
sary to measure their clinical practice. There is some evidence from the medical
literature that skill training in addition to confidence building is a major deter
minant of whether clinicians attempt new behaviours and persist in the face of
obstacles. It may not be well understood that carrying out standardized mea
surement procedures requires training and practice.
The collection of accurate and objective (i.e. measurement-generated) data about
an individual's motor performance in critical everyday actions and documenta
tion of the specifics of intervention makes a critical contribution to developing
best possible practice. The measurement of outcome without documentation of
intervention details has, however, limited relevance.
Objective information is critical to the design and ongoing modification and
variation of a training program. The information collected provides the know
ledge necessary to plan appropriate exercise and practice, and can be used to
collate data on particular patient groups within a centre and between centres
or countries. Collecting such information is not only an important means of
establishing best possible practice but also of making changes to practice as
more effective methods of intervention are developed.
Measures used may involve both functional and laboratory tests depending on
the questions to be answered and the availability of equipment and technical
expertise. Data are collected on the individual's entry to physiotherapy, at regu
lar intervals during rehabilitation, and on discharge and follow-up. Follow-up
measurements are necessary to provide information about real outcomes, such
as retention of gains after discharge from rehabilitation and how an individual
functions at home and in the community. Some suitable measures of health
needs and status are described below.
Many valid and reliable measures exist for evaluating change in functional
performance. These are listed in each chapter after the guidelines. Tests for
sensation, muscle strength, cardiovascular responses and range of motion
are also included. Standardized protocols describing how the test is adminis
tered must be observed during initial testing and repeated on subsequent
testing. The examination of similarities in test characteristics, and on ceiling
and floor effects will assist the future selection of a core group of clinically use
ful measures to monitor change in motor performance (e.g. Bernhardt et al.
Functional outcomes
Global instruments
Functional outcomes are commonly measured with global instruments such as
the Barthel Index (Mahoney and Barthel 1965, Wade 1992) or the Functional
Independence Measure (FIM) (Granger et al. 1986). These instruments mainly
focus on dependence (i.e. assistance required) and do not include social and
emotional aspects or the patient's perceived quality of life. These factors have
tended to be neglected areas of research.
Quality of life
The Sickness Impact Profile (SIP) (Bergner et al. 1981) is one of the most widely
used measures to assess quality of life (Qoi). The original form included 136
items. A short, stroke-adapted, 30-item version (SA-SIP30) has more recently
been developed and found to be a feasible and valid clinical measure to assess
patient-centred outcomes after stroke (van Straten et al. 1997). Sub-sections
included are: Body Care and Movement, Social Interaction, Mobility, Communi
cation, Emotional Behavior, Household Management, Alertness Behavior,
Ambulation. The Nottingham Health Profile is another short and simple meas
ure of health needs and outcomes (Hunt et al. 1985).
The Human Activity Profile (HAP) samples 94 acrivities across a broad range
of energy requirements (e.g. getting in and out of a chair or bed without assist
ance) to higher activity levels (e.g. vacuuming carpets for 5 minutes with rest
periods; walking for 2 miles non-stop). The items are based on estimated meta
bolic equivalents that allow quick and meaningful measurement of changes in
activity levels. Respondents answer each item with one of three responses 'Still
doing activity', 'Have stopped doing this activity', 'Never did this activity'
(Fix and Daughton 1988).
Patient satisfaction (self-efficacy) has also received little attention in rehabilita
tion. Input from patients can be sought about such issues as their expectations
and experiences during rehabilitation. Questionnaires completed anonymously
provide an indication of the individual's perception of the value of aspects
of rehabilitation such as their exercise and training programme. Although
patients are seldom asked to evaluate their experiences during rehabilitation,
there are several insightful publications written by individuals who have sur
vived a stroke (e.g. Brodal 1973, Hewson 1996, McCrum 1998, Smits and
Smits-Boone 2000).
In conclusion, there should be great satisfaction among clinicians knowing that
they are up-to-date in what is considered best possible practice and that they
have an ongoing commitment to pushing the limits of what patients can
achieve. All of this depends on an organized systematic plan for intervention
and on the collection of objective data, the results of which are more informa
tive and satisfying than observations that are subjective and open to bias. It is
encouraging to see a growing number of clinical investigations that record
quantitative change in movement parameters together with content, frequency,
dosage and intensity of physiotherapy intervention.
Figure 1.7 shows an outpatient circuit class in progress for the lower limbs in
the Physiotherapy Department at a New South Wales Hospital. Work stations
illustrated include walking sideways, standing balance, step-ups and stair
walking. In Figures 1.8-1.10 individuals are working at specific stations. They
attend the class for 6 weeks and functional performance is measured at the
beginning and completion of the programme with 6- and 12-month follow-up
testing (Fig. 1.11). A booklet is given to participants with details about the
class (Fig. 1.12) as well as a folder with instructions and diagrams of exercises
to practise at home and how often, with a form on which to record what they
have done. Photographs were taken by courtesy of F. Mackey and colleagues,
Illawarra Area Health Service.
28 Introduction
__ Overview of lower limb circuit class.
x 10 from bed
x 10 from bed
x 10 bed to chair
Feet back
Shoulders forward
Weight on both legs
(a) Sit-to-stand station. Holding a glass of water reinforces the need to balance. (b) The
line drawing illustrates the action and the number of repetitions to be performed.
Brain reorganization 29
FIGURE 1.9 Two other stations. Left: heels raise and lower with back against the wall. Right:
modified squats against the wall.
FIGURE 1.10 Participant's husband supervises practice at stairwalking station.
The form for data collection. (Courtesy of FMackey, Physiotherapy Department.)
(no ai9)
Interpreter required Y
(no aid)
12 month followup
of sessions J
Date of incident 1st eVA Y N
Relevant medical history/precautions
Lower Limb Circuit Class - Reassessment
5. Date of birth
6. Sex
9. practitioner
10. 10m walk (with aid)
11. Timed Up and Go test
12. ,10 steps - up
13. 6 minute walk
14. Step test (L)

Print name
Lower Limb Circuit Class - Assessment

walk.(With aid).
2. Timed Upand Go test
3. 10 steps - up
- down
- _ ..
f 4. 6 minute _.
5. Step test (L)
I6 month follow.up. -
r Signed
30 Introduction
Brain reorganization 31
FIGURE 1.12 Explanatory six page booklet. Read from left to right, top then bottom. (Courtesy of
FMackey, Physiotherapy Department.)
What is the Lower Circuit Class?
This is an exercise class which
aims to
a) improve your ability to stand-up,
balance and walk
b) improve your ability to walk up
and down stairs and
c) improve your fitness and your
ability to walk long distances
What will I do?
a warm-up where you will walk at a
comfortable pace for 5 to 6 min
stretches for your calf muscles.
a series of 8 workstations or exer
cises. You will exercise at 1 work
station for 3 to 5 minutes then move
onto the next workstation until you
have completed all 8 workstations.
Will I be disadvantaged if my
partner cannot attend?
No. We will endeavour to ensure that
you understand what to do at each
workstation and that you are able to
practice safely at home.
What should I wear?
You will need to wear comfortable
clothing and shoes. We usually recom
mend shorts or trackpants, a T-shirt,
walking shoes and a warm top for
occasions when we are outside.
and lastly, walking inside or walk
ing outside or on slopes, stairs and
When does the class run?
The class is run on a Monday and
Thursday at 9.00 am for six weeks
How many people will be in the
There will be somewhere between
4 to 8 participants in each class
with 2 staff members to supervise the
Can the hospital provide me with
The hospital is able to provide a
limited number of people transport on
our bus. On most occasions we will
endeavour to find alternative trans
port where possible.
Will I need to practice at home?
Yes. To maximize your recovery you
will need to practice at home. The
therapist will discuss with you what
exercises you are safe to continue
with at home. It is important to con
tinue the exercises you have been
given for home even when the class
has finished.
Are interpreters for available?
Yes. Interpreters are available for
people from most non-English speak
ing backgrounds.
Do I exercise on my own?
Yes. The 2 therapy staff will advise
you what to do and there will be dia
grams advising you what to do at
each station, then you will spend time
at each workstation exercising with
as well as without the therapist. We
feel this semi-supervised practice is
vital to help you learn how 10 keep
practicing at home.
Can my partner attend with me?
Yes. We welcome any person who is
interested in attending to assist you
in completing each station. This may
also help you when you are prac
ticing at home. There may be some
occasions when we will practice with
out partners present.
Eillisiotherapy Requirements
1. You are expected to arrive for
appointments on time.
2. If you are having difficulties arriving
for your appointment on time, or
are unable to attend, please notify
the Physiotherapy Department.
(Telephone 42238211) Failure to
attend on 2 occasions will lead to
automatic discharge and your
referring doctor will be notified.
Should you have any problems with
any part of your program, please dis
cuss this with the therapist.
Training guidelines
2 Balance 35
3 Walking 76
4 Standing up and sitting down 129
5 Reaching and manipulation 159
Biomechanical description
Balancing during motor actions in
Balancing during motor actions in
Balancing during body transport
Age-related changes
Analysis of motor performance
in sitting and standing
Research findings
Observational analysis
Guidelines for training
Sitting balance
Standing balance
Soft tissue stretching
Strength training
Functional tests
Biomechanical tests
Balance dysfunction, particularly in standing, is a devastating sequel of stroke
since the ability to balance the body mass over the base of support under dif
ferent task and environmental conditions is one of the most critical motor con
trol factors in daily life. Training balanced movement may be the most
significant part of rehabilitation.
Balance involves the regulation of movement of linked body segments over
supporting joints and base of support (MacKinnon and Winter 1993). It is the
ability to balance the body mass relative to the base of support that enables us
to perform everyday actions effectively and efficiently. Many of these actions
involve the lower limbs supporting and moving the body mass over the feet in
standing or in sitting during reaching and manipulative tasks, raising and lower
ing the body mass to stand up, or transporting the body from place to place
as in walking. Sometimes we just stand still. However, even simple actions
such as breathing and turning the head are characterized by oscillations in the
line of gravity which are countered by muscle activity and small, barely
detectable movements, usually at ankles and hips (Bouisset and Duchenne
1994, Vandervoort 1999). The ability to balance and maintain a stable posture
is integral to the execution of motor skills and cannot be separated from the
action being performed or the environment in which it is being carried out
(Carr and Shepherd 1998). The regulation of the dynamic interactions between
linked segments as they move one upon the other therefore plays an important
role in balance control (Nashner and McCollum 1985, Yang et al. 1990).
The mechanical problem of remaining balanced as we move around is a particu
lar challenge to the central nervous system (CNS). Ghez (1991) refers to a
family of adjustments needed to maintain a posture and to move which has
three goals:
to support the body against gravity and other external forces
to maintain the centre of body mass aligned and balanced over the base of
to stabilize parts of the body while moving other parts.
These postural adjustments to preserve or regain balance are brought about by
muscle activations and movements of segments. They are the consequences
both of internal mechanisms (e.g. muscle strength, visual, tactile, proprio
ceptive, vestibular sensory inputs) and of the environment in which action
takes place (Slobounov and Newell 1993). In standing, the CNS functions
to keep the centre of body mass (CBM) within the limits of stability, which
is the perimeter of the base of support beyond which we cannot preserve bal
ance without making a new base of support (e.g. taking a step) or falling.
Whether or not we take a step appears to be governed not only by the position
of the CBM relative to the base of support but also the horizontal velocity of
the CBM as the body moves about over the base of support (Pai and
Patton 1997). Our perceived limit of stability may differ if we anticipate or
sense a threat to balance, misinterpret visual inputs or experience fear or
The roles of sensory inputs are controversial but it is apparent that the inputs
critical to the events unfolding are coordinated in an action- and environment
specific manner. Redundancy in the sensory system allows for compensation
when one system is dysfunctional (Winter et al. 1990). Visual inputs, however,
appear particularly important in balance control, providing information about
where we are in relation to our surroundings and enabling us to predict
upcoming perturbations. For example, vision enables us to judge when we will
arrive at the kerb so we can time our stride for appropriate foot placement
(Patla 1997). It assists us to orient ourselves vertically, and seems particularly
critical with a small support base or a low level of skill (Slobounov and Newell
1993). Cutaneous inputs from the soles of the feet may be critical to our ability
to balance in standing and for the control of stepping (Do et al. 1990). Paying
attention to the relevant information is also important, particularly so when
balance presents a challenge (Gentile 2000). If attention is divided between
two tasks this may present a problem to an elderly person (Chen et al. 1996)
and to someone with poor balance after a stroke. The ability to sustain atten
tion to a task and avoid distraction appears to be related to the ability to bal
ance after stroke (Stapleton et al. 2001).
As we go about our daily activities, the postural system must meet three princi
pal challenges: maintain a steady posture, generate postural adjustments that
anticipate goal-directed movements and are adaptive as movements unfold,
and react swiftly and appropriately when we predict a threat to balance or
when an unexpected event perturbs our balance. The functional actions we
engage in during daily life make different ongoing mechanical and functional
demands on the segmental linkage according to the position we are in (body
alignment) and the requirements of the task. Reacting to externally imposed
destabilizing forces (standing on a moving walkway, being pushed in a crowd)
and transporting our body mass from one position to another (walking, stand
ing up) make different demands. Hence the patterns of muscle activity and seg
mental movement (what joint movement occurs, what muscles act as prime
movers or synergists) differ also. [n other words, postural adjustments are both
task and context specific, a finding which is replicated in many laboratory and
clinical studies (Dietz et al. 1984, Horak and Nashner 1986, Do et al. 1990,
Eng et al. 1992).
Throughout all the actions in which we are skilled we therefore preserve bal
ance as part of that skill. We fall only when an unexpected event occurs and
we fail to respond quickly enough when tripped up by an unseen or misjudged
obstacle, or when we stretch ourselves to the limits of performance capability
and 'over-balance'. The likelihood of falling increases in the weak or frail, or
in the presence of an impairment affecting the neuromuscular or sensory
systems, since under these conditions actions which are normally performed
skilfully can no longer be performed effectively.
Major requirements for good balance are an accurate sense of being balanced,
the ability of muscles, particularly of the lower limbs, to produce force rapidly
and at the appropriate time, and muscles which are extensible, i.e. not stiff or
short. The systems involved need to be adaptive, since balance control requires
the ability to adapt our movements to changes occurring both internally and in
our external environment.
The adjustments we make to preserve balance are flexible and varied. It is there
fore clear that the terms 'equilibrium reactions', 'balance reactions' and 'pos
tural reflexes', introduced into physiotherapy from the scientific literature half a
century ago and still in use for the testing of balance and as a goal of therapy
(de Weerdt and Spaepen 1999), do not account for balance control during
skilled, purposeful movements. These terms, which reflect an older view that
balance is essentially reflexive or reactive, can divert attention in physiotherapy
away from the specificity of postural (balance) adjustments and the fact that
they occur in anticipation of and during voluntary motor actions as well as in
response to a perturbation. Patients need to acquire again considerable skill in
preserving balance under conditions encountered in daily life, a large propor
tion of which involve intentional movement of the limbs and the body mass.
Regaining the ability to perform any functional action following a neural lesion
is almost entirely a matter of learning to balance body segments over the base
of support, yet intensive and varied balance training is not always the major
priority in clinical practice that it should be. It is highly likely from the evidence
so far that patients will only improve if they practise a range of actions under
the conditions occurring in daily life and if they are challenged to extend the
limits of their stability. From the research evidence so far, it is also critical that,
when lower limb muscles are too weak to support the body mass and to regulate
movements of the body mass, training also involves intensive muscle strength
ening exercises for the lower limbs.
Forces which disturb balance derive from the effects of gravity (gravitational
forces) and from the interactions between segments which are generated as we
move (reactive, interactive forces). Hence, gravitational and acceleration forces
must be controlled for posture and equilibrium to be maintained. Below is a
brief description of the principal findings from investigations of balance in
standing and sitting, and during walking, standing up and sitting down.
Balancing during motor actions in standing
In quiet stance, small movements of the body mass occur, making up what is
called postural sway. The extent and direction of sway is measured at the sup
port surface by calculating the movement of forces under the feet, indicating the
centre of pressure (COP). The amount of body sway in quiet stance is affected
by such factors as foot position and the width of base of support (Kirby et al.
1987, Day et al. 1998, Gatev et al. 1999).
The measurement of postural sway gives no information about the movements
occurring within the segmental linkage above the base of support (Kuo and
Zajac 1993). Although postural sway is sometimes used as a measure of bal
ance in clinical trials, it should be noted that the extent of postural sway is not
a sufficient measure of functional postural stability (Horak 1997). Able-bodied
individuals, including ballet dancers and athletes, may have a relatively large
extent of sway (Horak 1987, Massion 1992), while highly skilled pistol shoot
ers (Arutyunyan et al. 1969) have a minimal degree of sway as they prepare to
pull the trigger.
Effects of task and environment on balance mechanisms
Studies of self-initiated movements of the arm performed in standing show that
postural adjustments occur before the action as well as during the movement
(Eng et al. 1992). These anticipatory or proactive postural adjustments (noted
from electromyography, force plate and motion analysis) minimize the antici
pated destabilization of the body caused by the action being performed
(Zattara and Bouisset 1988). Anticipatory and ongoing postural adjustments
ensure the body's centre of mass remains within the base of support. It is of
particular note that in arm raising tasks the potentially destabilizing effects on
the CBM caused by the arm flexing forward are counterbalanced by appropri
ate leg muscle action (Belenkii et al. 1967, Lee 1980, Cordo and Nashner
1982, Horak et al. 1984, Nardone and Schieppati 1988). These findings illus
trate the interactive effects that occur in a segmental linkage.
Postural adjustments are an integral part of an action since they are highly spe
cific to such factors as initial body position (Cordo and Nashner 1982), condi
tions of support (Nashner 1982), speed and amplitude of movement (Horak
et al. 1984, Aruin and Latash 1996). The experiments illustrated in Figure 2.1
clarify the specificity of muscle activity and illustrate the potent effects of exter
nal support. Subjects interacted with a handle in four different ways (Nashner
Balance 39
FIGURE 2.1 Task- and context-specific coordination of arm (biceps brachii) and leg (gastrocnemius)
muscles with different tasks performed in standing with subject holding a handle.
(A) Subject pulls on handle. (B) Unexpected movement ofthe handle away from the
subject, with chest support. (C) As in B but subject free standing. (D) Unexpected
platform movement forward. (Adapted from Nashner 1983, with permission.)
(A) Subject pulls (B) Handle pulls - (C) Handle pulls - (0) Platform forward
on handle shoulder braced shoulder not braced
500 msec
o 500 o 500 o 500 o
f 0 f f f
Tooo r ~
1982): (A) Unsupported in standing, the subject pulled on the handle. This chal
lenge to balance resulted in activation of a postural leg muscle (gastrocnemius)
prior to the pull and before the activation of the arm muscle (biceps brachii).
(B) With support provided at chest height, when the handle was unexpectedly
pulled away from the subject, minimal leg muscle activity occurred since main
taining a balanced position actively was not necessary. (C) When this support
was removed, however, calf muscle activity was again evident.
It is not always lower limb muscles that playa balance preservation role in
standing. When subjects held the fixed and stable handle and the support sur
face was unexpectedly moved forward (D), the arm muscle turned on before a
small contraction of the leg muscle. That is, arm muscle activation was crucial
in preventing balance from being lost since the hand was grasping the only
object that provided stability. It was noted that the amplitude of leg muscle
activity was also markedly reduced when subjects touched a rail with one fin
ger. A similar result was reported in the study of rising on toes (Nardone and
Schieppati 1988), with a reduction in leg muscle activity when subjects held on
to two handles.
A group of actions has been studied in which the body mass is shifted in order
to free one leg from support. From the evidence, these actions appear to share
similar biomechanical characteristics. They all involve raising one leg in stand
ing and include studies of single leg raising (Rogers and Pai 1990), gait initi
ation (Kirker et a!. 2000) and lifting one foot on to a step (Mercer and Sahrmann
1999). In these actions, the CBM is displaced laterally over the stance limb,
40 Training guidelines
toe off
' ~
R Adductor
Phase of gait Initial lateral Left toe Left heel
initiation weight transfer off strike
L Adductor
Lateral torque
R Gluteus
L Gluteus
Stepping with left leg
Muscle activation plays balance as well as primary movement generation roles.
Such is the task- and context-specific nature of movement control, these roles
merge into one another when movement is well controlled. Another group of
studies investigating functional tasks performed in standing demonstrate the
complexity of this control. Studies of forward and backward trunk bending
(Thorstensson et al. 1985, Crenna et aJ. 1987), rising on to toes (Lipshits et al.
(a) Subject standing on force-plate, raises R leg sideways. The centre of pressure
(COP) first moves toward the R (upper peak) in order to move the centre of gravity
(COG) to the L over the supporting leg. The COP then moves to the L (lower peak)
under the supporting foot. Note that when the COP completes its movement to the L
(at T
), the ankle is raised sideways. (Adapted from Lee et al. 1995, with permission.)
(b) Averaged MG for an able-bodied subject stepping with the L leg for gait
initiation. Before stepping with the L leg, weight is moved to the R by L GM and R
ADD. As weight is transferred to the R leg, R GM and L ADD activity peaks (ADD,
adductor; GM, gluteus medius). (Adapted from Kirker et 01. 2000, with permission.)
allowing the other foot to be lifted. This frontal plane movement is itself com
plex. The data show that the centre of foot pressure is shifted toward the leg to
be raised milliseconds before the CBM is shifted toward the single support leg
(Fig. 2.2a). This shows that weight is shifted to the support limb partly by muscle
activity of the opposite leg (Fig. 2.2b). The point to be noted is that both lower
limbs are involved in any shift of body mass in a lateral direction (Jenner et al.
1997, Holt et al. 2000). This action requires both propulsive force to move the
body mass sideways and braking force to ensure it does not move beyond the
limit of stability (Rogers et at. 1993). Postural adjustments involved in lateral
movements of the body mass occur principally at shank-foot (invertors, evertors)
and thigh-pelvis (abductors, adductors) linkages (Winter 1995). Figure 2.2b
illustrates the action of hip abductor and adductor muscles in stepping with the
left leg (Kirker et al. 2000).
Ankle :/Moving side
Shoulder I f--+
. ~ : Support side
Ip I ~
Centre of t '. I
1981, Nardone and Schieppati 1988) and bending down to pick up boxes of
different weights (Commissaris et al. 2001) found anticipatory postural adjust
ments which served to reposition and stabilize the CBM prior to the primary
action and were clearly part of the action itself.
Investigations of rising on to toes in standing (Lipshits et al. 1981, Nardone
and Schieppati 1988) have shown preparatory activation of tibialis anterior
prior to the heel raise associated with forward displacement of the body mass,
followed by soleus, gastrocnemius and quadriceps contraction. Contraction of
calf muscles raises the heel. Contraction of quadriceps counteracts the flexing
force applied at the knee by the two-joint gastrocnemius, and maintains the
knee in extension. These studies illustrate the complexity of synergistic muscle
activations in what appears to be a rather simple action.
When subjects flex their lower limbs to lift a box from the floor, examination
of anteroposterior ground reaction forces (GRFs) has shown that anticipatory
adjustments occurring prior to the lift were adjusted to take account of the
weight to be lifted and to optimize the lifting phase (Commissaris et al. 2001).
Findings such as these show how the different constraints imposed by task and
context are adjusted for by the motor control system and point out the need to
include a variety of actions with differing constraints in training protocols.
Postural responses to unexpected perturbations
These have been studied principally in regard to support surface perturbations,
an environmental condition increasingly encountered in modern life. Studies
such as these, and others which have examined stumbling during walking, pro
vide information about the mechanisms of reactive balance responses which
form the emergency back-up system when our predictive mechanisms fail (Patla
1995, Huxham et al. 2001). Findings have demonstrated rapid responses to
unexpected platform movement which were directionally specific and focused
on distal musculature linking shank and foot. Even these rapid responses were
adaptive to context.
Putting out one or both hands is a typical reaction to loss of balance. However,
in standing, taking a step is used as a means of ensuring stability when we per
ceive a risk of falling (Maki and McIlroy 1997). This compensatory stepping
appears to occur not only in response to loss of balance, but as a preferred
strategy, even when perturbations are relatively small (Maki and McIlroy
1998). Stepping is initiated very rapidly, and inability to step accurately and
quickly enough may be a major cause of loss of balance confidence and of falls
in at-risk individuals. A recent study of older individuals reports that slow speed
of stepping plus delayed gluteus medius muscle onset in the stance limb may
predict the likelihood of falls (Brauer et al. 2000).
Balancing during motor actions in sitting
In sitting, the CBM is closer to the base of support which is relatively large com
pared to standing. Effective performance of actions such as reaching for an
object involves not only the ability to maintain the seated position on seats of
various dimensions, but also to balance the body mass over the lower limbs
while performing a variety of actions in the frontal and sagittal planes, often at
more than arm's length. The lower limbs play an important role in stabilizing the
trunk and pelvis (Son et al. 1988). Although there has been relatively little inves
tigation of sitting balance, it is clear that, as in standing, postural adjustments
(balance mechanisms) are highly specific to task and environmental conditions.
The major role of the lower limbs in supporting and balancing the body mass
in sitting is evident from experimental studies of reaching beyond arm's length,
both at fast and slow speeds (Crosbie et al. 1995, Dean et al. 1999a,b). Lower
limb muscles are involved in anticipatory, ongoing and reactive postural
adjustments. Tibialis anterior is activated and ground reaction forces through
the feet occur in advance of and during fast arm movement. Calf (soleus) and
knee muscle activity (vastus lateralis, biceps femoris) brakes the forward
movement of the body mass towards the end of a reach (Crosbie et al. 1995)
(Fig. 2.3). The shank and thigh muscles playa less active role when reaching is
self-paced and within arm's length (Dean et al. 1999a,b).
In general, trunk muscles act as postural stabilizers during movements in sitting
that involve the upper limbs. However, when an action such as reaching involves
moving beyond arm's length in order to transport the hand to the target, the
trunk also plays a role in extending the distance that can be reached. That is,
movement of the trunk by flexion and extension at the hips is an integral part of
the reaching movement (Kaminski et al. 1995, Dean et al. 1999a,b). The distance
we can reach is also affected by the extent of thigh support (Dean et al. 1999b).
Reaching sideways in the frontal plane would be expected to be more destabil
izing than reaching forward in the sagittal plane since the perimeter of the base
of support is reached earlier. However, no investigations of movements in this
direction were found.
The upper body can move about much less when the feet are hanging free,
such as over the side of the bed, compared to when they are on the floor; i.e.
the limit of stability is reached much sooner with feet unsupported than with
feet supported. Hence there is less possibility for vigorous action (e.g. reaching
beyond arm's length) when the feet are unsupported (Chari and Kirby 1986).
In this position, muscles linking thighs to the pelvis (e.g. iliopsoas, rectus
femoris, gluteus maximus) would be critical. Whether or not the feet are on the
floor, muscles linking pelvis and trunk (e.g. spinal flexors and extensors)
ensure appropriate alignment and stability of the trunk and head during activ
ities performed in sitting.
Balancing during body transport
Variations in task and environmental context affect balance mechanisms dur
ing gait. For example, balance is more challenged when walking in a dark

-soo -400 -300 -200 -1000 100 200 300 400 500
Fast reaching to a target with the R arm. Trial from one subject showing typical E:MG
traces from ipsilateral vastus lateralis (VL), biceps femoris (BF), tibialis anterior (TA),
soleus (Sol) and anterior deltoid (AD) in a forward reach.
160'% arm length
Balance 43
45 degrees
room or crossing a street (Huxham et al. 2001); balance has to be preserved
when walking on sand, on ice, and on a moving surface. Vision plays a particu
larly critical role in providing inputs about environmental features, including
changes occurring within the environment which are essential to the formula
tion of predictive adjustments and avoidance strategies.
Given the specificity of postural adjustments, it is not surprising that control of
balance while walking differs from control of balance in standing. Although in
standing the goal is to maintain the centre of gravity (COG) within the base of
support, walking is essentially controlled instability. The COG moves forward
along the medial borders of each foot, oscillating less than 2 cm to either side,
~ s
resulting in the body being in a constant state of 'imbalance' throughout single
support (MacKinnon and Winter 1993) (Fig. 3.1). Stance in double support
phase is not particularly stable since neither foot is flat on the ground (Winter
et al. 1990). It is evident that control of foot position after foot placement is
critical for stability (MacKinnon and Winter 1993).
The regulation of dynamic balance of the upper body during stance and the safe
transit of the foot during swing phase (toe clearance and foot landing) present a
particular challenge to the CNS. The large upper body mass (head, arms, trunk)
consists of two-thirds of the total body mass and the CBM is located about
two-thirds of body height above ground level. Walking therefore requires com
plex control of the movement of the total body mass over the supporting feet.
Balance is particularly challenged in stepping over obstacles, since the clearance
of both lead and trail limbs has to be greater than in unimpaired walking.
Since the base of support during single support phase is narrow, frontal plane
balance in particular requires precise control (MacKinnon and Winter 1993).
Balance of the upper body over the weightbearing hip appears to depend
largely on control of the pelvic segment at the hips, particularly by abductor
movements at the hip acting on the pelvis (MacKinnon and Winter 1993), and
also requires coordination between pelvis and trunk. In the sagittal plane, tight
coupling between hip flexors and extensors and knee muscles during single and
double support help balance the upper body over the lower limbs. As well as
the requirements of balance, the lower limbs must also be able to support the
body weight. Therefore, due to the effects of gravity, there is always net exten
sor activity in supporting limbs in order to prevent collapse, even when one or
two joints are flexing (Winter 1995).
Loss of balance may occur due to such factors as stumbling, tripping or slip
ping while taking a step, staggering while turning (Tinetti et al. 1986), and
when increasing speed. Precise control of the swinging foot, with toe clearance
of around 1-2 cm, plus control of all the segments involved in the stance and
swing of legs ensures we seldom trip. The process of turning consists of decel
erating forward motion, rotating the body (either by spinning on one foot or
taking a step) and striding out into a new direction, movements which can
cause staggering if balance is not controlled (Hase and Stein 1999).
Standing up (STS) and sitting down
In STS, the control of linear momentum of the body mass, particularly in a
horizontal direction, plays a critical role in dynamic stability and requires com
plex interactions between accelerating and decelerating muscle activations.
There is evidence that healthy elderly individuals may decrease the speed of
standing up in order to decrease the demands made by momentum on weak
muscles and on balance control. Important muscle groups for controlling hori
zontal momentum of the body mass include quadriceps and in particular gas
trocnemius-soleus since their activity on the shank has a braking effect on
forward movement (Crosbie et al. 1995, Scarborough et al. 1999).
Falls due to loss of balance are reported to be common in elderly individuals
during standing up and sitting down (Tinetti et al. 1988), actions which require
It is
~ t .
~ t
~ e
complex coordination of body segments to transport the body mass from over
one base of support to another while preserving balance. Patients with weak
ness of lower limb extensor muscles and dysfunctional motor control experience
difficulties with balance throughout STS, for example, controlling both horizon
tal momentum of the body mass prior to thighs-off and while sitting down.
In conclusion, it is clear from the scientific evidence so far that postural adjust
ments are specific to task (reaching for an object, walking across the room,
standing up), and context (body position, environmental features). Even rather
small changes in task and conditions can produce marked changes in patterns of
muscle activity (Mercer and Sahrmann 1999). This specificity underlies the
apparent lack of transferability from balance training in one task to perform
ance of another, mechanically different task. For example, training to improve
lateral weight shift in standing does not necessarily transfer into improved abil
ity to walk (Winstein et al. 1989).
Stroke results in impairments which affect the force-generating capacity of
muscles, the control of movement and therefore the performance of a range of
different actions. From the scientific research so far we can assume that an
acceptable degree of skill will only be regained by functionally specific
strengthening exercise for the lower limbs and motor training of actions in
their usual environments. The ability to balance is trained as an integral part of
each action and it is the practice of balancing the body mass throughout per
formance of the action that enables the individual to optimize performance
and regain skill.
The effect of the ageing process upon balance is complex and unclear. The
results of studies of elderly people can be confusing if they include individuals
with eNS and/or musculoskeletal problems. Elderly individuals are at greater
risk of falling and it can be assumed that balance deficiencies are a significant
cause. Postural instability in the elderly and a tendency to fall is associated with
reduced vision, including loss of sensitivity in peripheral visual field, reduced
peripheral sensation, vestibular dysfunction and slowing of muscle reaction time
(Stelmach and Worringham 1985, Woollacott et al. 1988, Lord et a1. 1991), the
latter reported for both tibialis anterior (Woollacott et al. 1986) and gluteus
medius (Brauer et a1. 2000). Woollacott and colleagues (1988) reported
increased co-contraction to stiffen lower limb joints in older compared to
younger adults which may be an adaptation to deficiencies in balance. In add
ition, elderly individuals are at risk of developing pathologies which accelerate
degeneration in neuromusculoskeletal systems (Horak et a1. 1989).
Postural stability is constrained by biomechanical factors associated with
decreased musculoskeletal flexibility and muscle strength. Leg muscle strength
has been reported to decrease by 40% and arm strength by 30% from 30 to 80
years of age (Asmussen 1980). At ages higher than 70 years, muscle atrophy
appears to accelerate (Danneskiold-Samsoe et al. 1984). Such declines in
46 Training guidelines
strength impact negatively on balance in standing and during walking, stair
climbing and standing up. Risk factors for falling most commonly identified
are loss of muscle strength and flexibility (Gehlsen and Whaley 1990), particu
larly loss of quadriceps strength (Scarborough et a!. 1999). An association has
been found between a history of falls and weakness of ankle dorsiflexors and
plantarflexors (Whipple et a!. 1987), which may illustrate the critical role of
muscles that link the shank segment to the base of support, i.e. the foot.
Individuals with a history of falls have shown significantly decreased ability to
maintain one leg stance with and without eyes open (Johansson and Jamlo
1991), and significantly weaker lower limb extensor muscles compared to non
fallers (Gehlsen and Whaley 1990, Scarborough et a!. 1999).
Limited ankle joint mobility may provoke tripping and falling. Decreased
mobility may be due to increased passive resistance of the elastic component in
soft tissues (Vandervoort 1999) and muscle shortening (Mecagni et al. 2000).
A study of elderly individuals found loss of ankle mobility, particularly evident
in women, associated with stiffer connective tissue and reduced strength of
dorsi- and plantarflexors (Vandervoort et a!. 1992). Stretching and exercise
programmes have shown beneficial effects on ankle flexibility, with increases in
strength and decreases in stiffness (Vandervoort 1999).
Increased amplitude of body sway in quiet standing in older adults has been
commonly reported but the functional relevance of this is unclear. Tests of sway
in standing are not a direct measure of functional postural stability (Horak
1997). However, the results of one study of postural sway suggest that elderly
individuals may have difficulty preserving balance in standing when they turn
their head (Fig. 2.4) (Kocej2 et a1. 1999). Very elderly individuals hospitalized
for various reasons and who had poor balance in standing were reported to
Postural sway (mean sway in a sagittal direction) in quiet standing and during a
volitional head turn with eyes open in a group of elderly compared to young
individuals. (Adapted from Koceja et al. 1999, with permission.)
2' 7
2 (/)
Young Elderly
consistently lose their balance in a posterior direction (Bohannon 1999). This is
a common observation in elderly people after a period of bedrest and seems to
reflect a perceptual bias away from the erect position which may be provoked
by the supine position. Older people seem also to have difficulty controlling lat
eral stability, noticeable during stepping to regain balance, which is a particular
risk in the elderly as it is associated with hip fracture (Maki and McIlroy 1997,
There is evidence that a programme of lower limb strengthening exercises can
have a positive effect in the elderly on muscle strength, balance and gait
(Fiatarone et al. 1990, 1994, Johansson and Jarnlo 1991, Lord and Castell
1994, Means et al. 1996, Campbell et al. 1997, Krebs et al. 1998, Hortobagyi
et al. 2001). There is also evidence that a strength and balance exercise pro
gramme can improve other motor functions involving balance and result in a
reduction of falls in the very elderly (Shumway-Cook et al. 1997). One study
of women (Campbell et al. 1997), that consisted of open chain (with ankle cuff
weights) lower limb strengthening exercises plus balance exercises, showed
that at 6 months the women had significantly improved their balance and their
ability to stand up compared to the control group. The exercises were done for
30 minutes, at least three times a week, and the programme included practice
of tandem standing and walking, walking on toes and heels, walking back
ward, sideways and turning around, stepping over an object, bending to pick
up an object, stair climbing and knee squats. After 1 year the exercise group
had a lower rate of falls than the control group, 42% of them were still doing
the exercise programme more than three times a week, whereas the control
group had become less active and their fear of falling had increased.
Frequent, intensive, stimulating and enjoyable strength and balance training
classes may help elderly stroke patients develop confidence and an understand
ing of the need to keep active. From the evidence so far, it seems clear that exer
cise programmes need to go on for several weeks and that individuals will need
to continue exercising in order to maintain the gains achieved (Vandervoort
Other exercise programmes emphasizing sensory function (Telian et al. 1990,
Ledin et al. 1991, Hu and Woollacott 1994) have been successful in improving
aspects of balance such as the ability to stand on one leg. Tai Chi Chuan train
ing, which emphasizes control over large movements of the CBM, has also
been found to have positive effects on functional balance, decreasing fear of
falling and frequency of falling (Tse and Bailey 1991, Wolf et al. 1997, Wong
et al. 2001). This training combines slow and gentle movements which flow
into each other, and includes stepping from one leg to the other with slow and
deliberate foot placement (Kirsteins et al. 1991).
It is evident from the research available that the consequences of physical
inactivity can be very serious in elderly individuals, with sequelae that include
loss of strength, stability, cardiovascular fitness and joint flexibility. A major
challenge for those responsible for physiotherapy after stroke is to ensure that
disuse effects do not compound the functional problems resulting from an
individual's stroke. Many people who have a stroke, and it is highly probable
that this applies to frail elderly women in particular, may have, before their
stroke, developed musculoskeletal disuse effects, poor balance, loss of self
confidence and fear of falling (Tinetti et al. 1994).
The sensory and motor impairments after stroke which impact upon balance
may include muscle weakness, decreased soft tissue flexibility, impaired motor
control and sensory impairments. The functional results of the motor control
impairments are loss of coordination, loss of an accurate sense of being bal
anced, and therefore an increased risk of falling. In addition to abnormalities
of balance, posture (the ability to take up and maintain a specific segmental
relationship) may also be affected in some individuals. Difficulty achieving and
sustaining an appropriate posture (segmental alignment) may be noticeable in
the patient's initial attempts to get into and maintain the sitting or standing
position and in those individuals with an impaired sense of their body position
in relation to the vertical.
Reduced muscle function, in the lower limbs in particular, affects the ability to
support, shift and balance the body mass. However, it is not only the inability
to generate appropriate force but also difficulty initiating, timing and sequen
cing muscle forces, sustaining force, and generating force fast enough which
affect the ability to balance (di Fabio 1997). These elements of dysfunction
affect anticipatory as well as ongoing postural muscle activity, with muscles
activated later than in the able-bodied (Viallet et al. 1992, Diener et al. 1993).
Delay in initiating muscle activity and slowness to build up force affect both
the preparation for an impending disturbance to stability and the speed of
response to loss of balance. Consider, for example, the destabilizing effects of
raising the arm in standing if preparatory muscle activity and joint movement
do not take place prior to the destabilization. A particular difficulty for many
patients is a decreased capacity to take a step when in danger of falling in
standing or in walking.
Secondary muscle adaptations, such as length and stiffness changes, also have
a significant impact on muscle activation and balance. Decreased soft tissue
elasticity and muscle shortening limiting ankle joint mobility may have a par
ticularly strong effect following stroke as they do in the elderly (Vandervoort
1999). Sensory system and perceptuo-cognitive dysfunction also impact nega
tively on balance in sitting and standing, particularly in the early stage after
stroke, and may include disordered somatosensory, labyrinthine, visual func
tion and perceptuo-cognitive deficits such as unilateral spatial neglect. One
cause of difficulty in stepping out for older individuals is loss of cutaneous
sensation from the plantar surface of the foot.
Falling has been reported as a frequent occurrence in individuals following
stroke and after their discharge from hospital, particularly those who are
~ s
60 years of age or over. One study reported that 73% of the 108 individuals
studied had a fall in the 6 months following discharge. Of the 108 individuals,
21 % reported having fallen prior to their stroke. Most falls occurred while
performing everyday activities such as walking, standing up and extreme
reaching, activities which in particular challenge an unbalanced system. Most
individuals had difficulty getting up from the floor when they fell, including
those who had been trained by a physiotherapist in how to do so. Individuals
who had fallen at least twice were depressed and less active socially (Forster
and Young 1995).
Research findings
Some biomechanical analyses of balance dysfunction in standing following
stroke have focused on laboratory investigations of responses to unexpected
perturbations, such as support surface movements or lateral pushes. Studies of
support surface movement have shown abnormal coordination of postural
responses (Horak et al. 1984, di Fabio and Badke 1990, di Fabio 1997), with
latency of postural muscle response favouring the non-paretic limb, reductions
in magnitude of response, including absence of any response from some muscles,
and agonist-antagonist coordination deficits.
Studies of the ability of stroke patients to withstand a force applied laterally to
the hip region to either side have demonstrated slowness in muscle activation,
difficulty with sustaining muscle force to resist displacement and also in coping
with the release of the applied force (Lee et al. 1988, Wing et al. 1993, Holt
et al. 2000, Kirker et al. 2000). These studies demonstrate ways in which indi
viduals adapt to severe muscle weakness by using stronger muscles. For example,
during sideways pushes in standing, activity in adductor muscles of the stronger
leg increased to compensate for diminished and delayed recruitment of hemi
paretic abductor muscles (Jenner et al. 1997).
Force platform studies of postural sway during quiet standing following stroke
have demonstrated both increased and decreased sway (Mizrahi et al. 1989).
The relevance of such results is not clear. Measures of postural sway are not
necessarily correlated with tendency to fall or with an individual's perception
of stability (Horak 1997), postural sway is poorly related to dynamic actions
such as walking (Gill-Body and Krebs 1994, Colle 1995), and limiting body
sway does not necessarily lead to improved stability.
In outcome studies of individuals with brain lesion, no relationship has been
found between decreased postural sway and functional scores or scores on
other balance tests (Fishman et al. 1997). Holding the body stiffly is a com
monly observed adaptation to poor balance and fear of falling after a brain
lesion (Carr and Shepherd 1998), and, therefore, elderly individuals and those
with neurological impairments who have minimal postural sway may not be
stable and are likely to fall. As several authors have pointed out (Horak 1997),
immobility (i.e. decreased sway) should not be confused with stability. In clini
cal interventions, the focus should not be on postural sway but on regaining
the ability to move freely without losing balance.
Force platform investigations of weightbearing characteristics confirm the clin
ical observation that less weight is taken through the affected lower limb in
standing and that there may be minimal movement towards that leg (Dickstein
et al. 1984, Goldie et al. 1996a). Studies of the ability of patients to deliber
ately redistribute their weight in different directions have shown a restriction
in the excursion of the COP laterally in both directions, backward and for
ward, with extraneous balance activity biased toward the non-paretic lower
limb (Dettman et al. 1987, di Fabio and Badke 1990, Goldie et al. 1996b,
Turnbull et al. 1996).
Difficulty shifting weight from one leg to another is particularly evident in studies
of taking a step or raising one leg. The findings typically show a reduced and
inappropriate displacement of the body weight to both sides, whether moving at
average or fast speeds (Rogers et al. 1993, Pai et al. 1994, Laufer et al. 2000). For
example, in making fast single leg raises, a delay in initiating ground reaction
force changes under the paretic limb was evident with a reduction in the propor
tion of actively generated horizontal force for propelling the body mass laterally
(Rogers et al. 1993). Studies of obstacle crossing show the difficulties patients
have with balancing on one leg while stepping over an obstacle (Said et al. 1999).
Difficulty balancing in sitting has been shown to be associated with poor func
tional outcome after stroke (Loewen and Anderson 1990, Sandin and Smith
1990). A study of individuals more than 1 year post-stroke has illustrated that
the inability to use the affected lower limb for balancing limits the distance
which can be reached and the speed of the reaching movement (Dean and
Shepherd 1997). Figure 2.5 illustrates the changes that took place in vertical
ground reaction forces after a period of training.
Following stroke, patients with muscle weakness and poor control lack effect
ive anticipatory, ongoing and reactive postural adjustments and therefore
experience difficulty performing actions which involve:
supporting the body mass over the paretic lower limb
voluntarily moving the body mass from one lower limb to the other and
one position to another
responding rapidly to predicted and unpredicted threats to balance.
As a result, in actions carried out in sitting and standing, and others including
walking, standing up and sitting down, the patient favours the stronger, better
coordinated and more rapidly responsive leg, and the ability to carry out even
simple tasks is seriously limited. Additional factors that are likely to impact
on stability include abnormal perception of verticality and other disorders of
visuospatial perception (Dieterich and Brandt 1992) and somatosensory per
ception (Niam et al. 1999).
Observational analysis
There is a clear distinction between measurement, i.e. the collection of objective
data, and observational assessment which is non-standardized and subjective.
Balance 51
FIGURE 2.5 Patterns of vertical ground reaction forces (GFR) (% body weight) of one stroke
subject, reaching in three different directions, pre- and post-training. Heavy line,
paretic leg; light line, non-paretic leg. Lower trace shows a healthy elderly subject
reaching with the R hand for comparison. Heavy line, L leg; light line, R leg. Dotted
n vertical lines indicate onset and end of hand movement. (Reproduced from Dean and
r:' Shepherd 1997, with permission.)
Ipsilateral Forward Across
. .
. .
L ~
. .
. .
1l -A
0 2 4 6 0 2 4 6
Time (s) Time (s)
. .
. .
0 2 4 6
Time (s)
Although it is not at present possible for biomechanical tests to be performed
in most clinics, there are many reliable and valid tests which give objective and
functionally relevant data about progress and outcome. Using these tests, each
patient is measured at the start of intervention, at intervals during rehabilita
tion, before discharge and at follow-up some time after discharge, in order to
obtain data about progress and final outcome.
Therapists must rely on visual observation for ongoing analysis and as a guide
to intervention. However, observational analysis has some major drawbacks.
For example, the validity of the observations and subsequent analysis are
dependent on the clinician's knowledge of normative biomechanics and skill in
interpretation. In addition, although such kinematic characteristics as segmen
tal alignment are observable, critical kinetic characteristics, such as ground
reaction forces and muscle forces, are not.
Biomechanical research is providing an increasingly large body of knowledge
about balance mechanisms, the findings of which assist the clinician in making
informed observations as the patient attempts to carry out tasks which range
52 Training gUidelines
from simple to complex depending on the patient's abilities. From these obser
vations, a skilled observer can make inferences about the effects of lower limb
muscle weakness, reduction in speed of muscle response, and lack of extensi
bility of soft tissues on posture and movement under specific conditions.
The taxonomy proposed by Gentile (1987) also assists the therapist's evalu
ation. It provides a clear picture of the varied conditions in which balance is
required (see Huxham et a1. 2001), reflects the environmental contexts in
which we perform daily actions, and makes clear the inseparability of balance
and movement in a gravitational environment. A person's movements need to
conform to the specific environmental features (the context) (Fig. 2.6) if they
are to be successful in achieving the goal (MagilJ 1998).
Given the variability of postural adjustments according to task and context, it
is not possible to provide detailed and critical biomechanical features of pos
tural adjustments in isolation from the tasks themselves. Balance control
involves bilateral interlimb and limb-upperbody-head coordination, with head
position being relevant to visual and vestibular inputs. In the clinic, the directly
observable characteristics indicative of balance in sitting and standing are the
relative alignment of body segments and such spatiotemporal adaptations as
speed of movement, step length, and width of base of support. It is possible,
therefore, to identify the inability of an individual to sustain the necessary body
alignment for stability and the spatiotemporal adaptations which are invariably
FIGURE 2.6 Reaching to pick up a glass involves limb and upper body movement including
postural adjustments that are specific to the shape of the object, its position and
what is to be done with it.
made in response to fear of losing balance. These are clearly observable and
form the basis of the therapist's analysis during training as the patient attempts
various tasks.
Body alignment
Inability to take up and maintain a specific segmental alignment (posture) in
sitting or standing may be evident when the person first attempts to sit or
stand. Normally when we are about to carry out an action we make certain
preparations to ensure effective performance. For example, we move to a posi
tion of postural (segmental) alignment that is more suited to the action, particu
larly to the organization of supporting balance activity (Magill 1998). This
postural alignment is one which favours the necessary muscle activity and feels
comfortable. When training patients, it is not 'sitting up or standing up
straight' that is important, but the taking up of a position which optimizes per
formance and maximizes success. Body alignment appropriate to quiet stand
ing and sitting is outlined in Box 2.1.
[BOX 2.1 Body alignment appropriate to quiet standing and sitting
Standing alignment Sitting alignment
Head balanced on level shoulders Head balanced on level shoulders
Upper body erect, shoulders over hips Upper body erect
Hips in front of ankles Shoulders over hips
~ e t a few cm (10cm) apart Feet and knees a few cm apart
Spatiotemporal adaptations
These include:
Changing the base of support
wide base of support (legslfeet apart, externally rotated)
- shuffling feet with inappropriate stepping
- shifting on to stronger leg (Fig. 2.7a,b).
Restricting movement of body mass
stiffening the body with altered segmental alignment (Fig 2.7a)
moving slowly
changing segmental alignment to avoid large shifts in COG
standing reaching forward - flexing at hips instead of dorsiflexing
ankles (Fig. 2.7c)
standing reaching sideways - flexing trunk sideways instead of moving
body laterally at hips and feet (Fig. 2.7a,b)
sitting reaching sideways - flexing forward instead of to the side
(see Fig. 2.9a)
in standing - not taking a step when necessary.
Using hands for support
holding on to support
- grabbing.
54 Training gUidelines
FIGURE 2.7 Adaptations made to avoid supporting and balancing body mass over the paretic
lower limb. (a,b) It is possible to reach sideways beyond arm's length by lateral
movement of the spine and increased shoulder girdle protraction. (c) In reaching
forward, reach distance can be increased by flexion at the hips and protraction of the
shoulder girdle. In both directions, however, the distance to be reached is limited.
(a) (b)
Balance 55
Balance cannot be trained in isolation from the actions which must be
relearned. In training walking, standing up and sitting down, reaching and
manipulation, postural adjustments are also trained, since acquiring skill
involves in large part the fine tuning of postural and balance control.
Postural adjustments are specific to each action and the conditions under
which it occurs. It cannot be assumed that practice of one action will
transfer automatically into improved performance in another. For example,
balancing exercises in sitting on an inflated ball, advocated in some texts,
focus on trunk and hip flexor muscles since the feet are not on the floor.
Transfer into improved sitting balance under different conditions and for
different tasks cannot be inferred. Exercises on a ball are difficult and
strenuous, and, since the patient may require considerable physical support
from the therapist, there may actually be little training effect.
Emphasis in the guidelines below is on performing tasks which require
voluntary movements of the body mass in sitting and standing. The
availability of harness support makes it possible for people to practise simple
balancing tasks in standing very early after stroke. It should be noted that an
overhead harness does not appear to interfere with responses to loss of
balance according to a test of able-bodied subjects (Hill et al. 1994).
Progressive complexity is added by increasing the difficulty under which
goals must be achieved, keeping in mind the various complex situations in
which the patients will find themselves in the environment in which they live,
both inside and outside their homes, and the precarious nature of balance.
As control over balance and confidence improves, tasks are introduced
which require a stepping response, and responses to external constraints
such as catching a thrown object and standing on a moving support surface
(e.g. moving footway).
The conditions under which balance control is necessary in daily life can be
broadly categorized as occurring:
during performance of a self-initiated action
when predicting destabilization and taking avoiding action
when making a reactive response as a last resort in an attempt to avoid
a fall.
The above categorization is a simplification. However, it provides a
framework for the development of training protocols, with balance trained
as an integral part of actions in the above categories. Practice is designed to
enable the individual to learn again, with a disordered system, how to
control movements of the body mass over the base of support and learn the
'rules' of preserving stability as part of acquiring skill.
Intervention aims to optimize balance by training:
balance of the body mass during voluntary actions in sitting, standing and
during body transport
quick responses to predicted and unpredicted destabilization.
For training to be effective given the impairments and adaptations following
stroke, it is also necessary to:
prevent adaptive shortening of lower limb soft tissues
increase lower limb extensor muscle strength and coordination
(for support of body mass).
Sitting balance
The ability to balance in sitting while reaching for objects both within and
beyond arm's length is critical to independent living. In the acute stage
post-stroke, major goals are to prevent medical complications associated
with the supine position and bed rest and to train balance in sitting.
Re-establishing sitting balance early is critical since it impacts positively on
many functions. It provides greater stimulus for gaseous interchange and
facilitates coughing, enables more effective swallowing. encourages eye
contact and focusing attention, communication and more positive attitudes,
stimulates arousal mechanisms, and discourages learned 'sick role' behaviour.
In contrast, mobility exercises performed in supine, such as rolling and
bridging, are often given excessive emphasis in early rehabilitation, taking up
time which should be spent with the patient upright and active.
The first exercises may be useful for patients who, early post-stroke, are
frightened of moving. The exercises take attention away from the need to
balance by focusing the patient's efforts on achieving a concrete goal.
Practising simple actions that involve small shifts of the body mass initially
enables patients to regain a sense of balance and confidence that they can
move independently. Actions are practised repetitively with no pause.
Head and trunk movements
Sitting on a firm surface, hands in lap, feet and knees approximately 15 cm
apart, feet on floor.
(i) Turning head and trunk to look over the shoulder, returning to mid
position and repeating to other side.
Ensure patient rotates the trunk and head, with trunk erect, and remains
flexed at the hips.
Provide visual targets, increasing the distance to be turned.
Stabilize the affected foot and prevent external rotation and abduction
at the hip if necessary.
Ensure hands are not used for support and that feet do not move.
(ii) Looking up at the ceiling and returning to upright.
Patient may overbalance backward and is warned to keep upper body
forward over the hips.
Balance 57
These exercises can also be done sitting over the side of the bed with
feet on a firm surface. A weak patient is assisted to turn on to the side
and to sit up over the side of the bed (see Fig. 8.1).
Reaching actions
Sitting, reaching to touch objects with the paretic hand: forward (flexing
at the hips), sideways (both sides), backward, returning to mid position
(Fig. 2.B). Very weak patients can practise reaching forward with arms
resting on a high table (see Fig. BAb).
When patient achieves a sense of balance, reaching with non-paretic arm
across body to load the paretic foot.
Reaching distance should be more than arm's length, involving movement
of the whole body, and as close to the limits of stability as possible.
Some exercises to train the ability to perform simple actions in sitting that require
small amounts of body movement: (a,b) Practising using the paretic L leg for support
and balance while reaching forward and sideways. (c) Moving the feet back
emphasizes the feet as an active part of the base of support and increases the load
through the paretic leg. The legs also playa part in moving back to upright which
involves lower limb extensor muscle forces and ground reaction forces, particularly
during faster movement. (d) At the first attempt at reaching (left) the person may be
reluctant to move to the paretic side; some improvement is evident (right) after a few
(a) (b)
58 Training gUidelines
_L_{_C_o_n_ti_n_u_ed_} _
In reaches toward the ipsilateral side, emphasis is on weightbearing
through the paretic foot since lower limb muscle activity is critical to
balancing in sitting.
Discourage unnecessary activity of non-paretic upper limb (e.g. raising
shoulder, grabbing for support).
Therapist can support affected arm during reaches to that side. Do not
pull on the arm.
If patient falls persistently toward the affected side, it may help the
regaining of a sense of balance if the patient is encouraged to reach
toward that side and return to mid position. The therapist can provide
Balance 59
FIGURE 2.9 Reaching sideways in sitting. The task has been modified by setting up the
environment to require less lateral movement of the body mass (centre) since he
cannot reach down to the floor (left). Weakness of the L lower limb muscles makes it
difficult to stabilize the leg. Reaching to the L gives him practice of loading that leg
and using it to shift his body moss bock to upright sitting.
(a) (b) (c)
some stability by anchoring the paretic foot on the floor. The effectiveness
of this strategy may result from the patient getting the idea of actively
controlling movement toward the affected side and back to the midline,
rather than compensating for the tendency to fall by keeping the weight
on the intact side.
This training can also be useful for patients with a misperception of the
true midline (the true vertical). The therapist should avoid pushing the
patient passively toward the midline.
Reaching to pick up objects from the floor, forward and sideways, one and
two hands. This can be made easier by placing the object on a box (Fig. 2.9).
A patient who lacks sufficient control of manipulation can reach to
touch the object.
If necessary, therapist can support affected arm. Do not pull on the arm.
Maximizing skill
The majority of patients trained with repetitive practice of these simple
exercises regain a degree of sitting balance within a relatively short period.
60 Training gUidelines
FIGURE 2.10 The task to practise is lifting the tray without spilling water or dislodging the utensils.
Varying practice to increase skill includes:
increasing distance to be reached
varying speed
reducing thigh support
increasing object weight and size to involve both upper limbs (Fig. 2.10)
adding an external timing constraint such as catching or bouncing
a ball.
Standing balance
Learning to balance in standing requires the opportunity to practise
voluntary actions in this position early in the acute stage. Standing balance
is not the static maintenance of a position. However, supporting the body
mass over the feet requires the ability to generate sufficient muscle force
in the extensors of the paretic lower limb to prevent the knee from
Opportunity to load the lower limb seems critical for promoting muscle
activity in that limb. Receptors sensitive to load have been shown to
activate leg extensor muscles during locomotion in cats. Sensitivity to load
has also been observed in humans, with a high correlation between degree
of leg extensor activity and percentage of body weight loading of that
Given that it is critical for the person to practise in standing, if the lower
limb tends to collapse, there are several ways of enabling balance practice.
Balance 61
FIGURE 2.11 (a) The knee support enables him to load the R leg without the knee collapsing while
he practises stepping forward. (b) The harness enables practice of actions in standing
which involve balancing the body mass without fear of falling.
(a) (b)
For example, a light splint prevents collapse at the knee, and a harness can
reduce body weight through the lower limbs. In standing, the patient
practises small movements of the centre of body mass. In addition, a
programme of exercises, with methods of eliciting muscle activity, increasing
strength and preserving length in critical lower limb muscle groups is
Asymmetrical standing is often a focus of the therapist's intervention.
Favouring one leg in standing is primarily due to inability to generate and
control supportive forces in the paretic limb and is an adaptation to the
problem of an unreliable or collapsing lower limb. Improving support
through the limb by strengthening exercises, and training in standing with
knee support or harness, should be a major focus in intervention (Fig. 2.11).
62 Training gUidelines
The exercises below include movements of the body mass ranging from
small displacements when patients are weak and apprehensive, progressing
to larger displacements performed faster. Focus is on actions such as turning
to see who is coming through the door or reaching to take an object. These
activities take the person's attention away from the need to balance by
directing it toward a concrete non-balance goal. This can be particularly
useful for a person early after stroke who is frightened of moving. Emphasis
from the start of training is on reaching beyond arm's length, varying object
placement and therefore direction and extent of reach, increasing
repetitions and complexity.
Head and body movements
Standing with feet a few cm apart, look up at ceiling and return to upright
(Fig. 2. 12a).
Correct tendency to fall back by a reminder to bring hips forward (hip
extension beyond neutral) before looking up.
Disallow foot movement.
Standing with feet a few cm apart, turn head and body mass and look
behind, return to mid position, repeat to other side (Fig. 2. 12b).
Ensure standing alignment is preserved, with hips extended while body
Disallow foot movement. If necessary put your foot against patient's
foot to stop movement.
Provide visual targets.
Reaching actions
Standing, reaching to take object forward, sideways (both sides),
One hand, both hands. Variety of objects and tasks. Reaches should be
beyond arm's length, challenging the patient to extend the limit of stability,
and return (Fig. 2.13).
Ensure movement of body mass takes place at the ankles and hips, not
just within the trunk.
Discourage a stiff posture and breath holding - encourage relaxed
Balance 63
FIGURE 2.12 Some exercises to train the ability to perform actions in standing that require
small amounts of body movement. (0) On her first attempt at standing, she
practises looking up at the ceiling to locate a target. She needs reminding to keep
:!se her body mass (hips) forward. (b) Looking around to locate a target on the wall to
the side.
(b) (a)
When these exercises are practised without a harness, therapist
should avoid holding the person as this diminishes the need to make
active anticipatory adjustments and corrections. Use of a belt with
grab handles (Fig. 2.14) may increase confidence of both patient and
Base of support is varied to increase difficulty (e.g. feet together, one
ot foot in front, one foot on a step, tandem standing). The patient learns to
control movement by reaching laterally to each side, forward and back
ward. Attention, however, is focused not on balance itself but on con
crete goals.
64 Training guidelines
FIGURE 2.13 (a) Reaching sideways, attempting to move weight over the paretic leg. (b) Training
balanced movement in a harness. Note (left) that he stands with more weight through
the R leg. He practises reaching to the L to pick up a cup (centre). Reaching across
the body (right) requires a greater shift to the L with some body turning.
Balance 65
FIGURE 2.14 A belt with grab handles gives some reassurance without interfering with the patient's
movements. (Handi-LiftIWalk belt, Pelican Manufacturing Pty Ltd, Osborne Park,
Western Australia 6017.)
Single leg support (with or without harness or splint)
Stepping forward with non-paretic limb to place foot on a step (Fig. 2.15).
Standing with either foot on step, practise reaching tasks.
Ensure that the stance hip extends. Exercise can be practised initially in
a harness.
Laufer et al. (2000) have shown that stepping forward on to a step of
different heights with the non-paretic leg significantly increases weight
borne through the paretic limb.
The exercise directs attention to the concrete goal of raising the leg
rather than the more abstract goal of shifting body mass.
66 Training gUidelines
FIGURE 2.15 Training stepping with non-paretic leg in a harness. (a) At first it is difficult for her to
keep her R knee extended to support her body mass. The therapist encourages her to
'move her hips forward' (extending her hip) and extend her knee. (b) Stepping with
her L leg on to a step. Therapist encourages her to move her body mass forward
toward the step, i.e. over the affected leg. (c) She no longer needs to hold on. Note
that as soon as she felt more stable she lets go of the handle as she performs a
prescribed number of repetitions.
(a) (b) (c)
Sideways walking
Walking sideways with hand(s) on wall (Fig. 2.16), or raised bed rail
(see Fig. 3.20a).
This exercise enables practice of shifting weight from side to side with hips
Picking up objects
Standing, lowering body mass to pick up or touch object, forward, sideways,
backward and return (Fig. 2.17).
Ensure that hips, knees and ankles flex and extend.
Start with object on stool to minimize distance to be moved (see
Fig. 2.17c).
Increase flexibility by changing base of support.
Balance 67
FIGURE 2.16 Walking sideways with some support. A line on the floor guides her to abduct/adduct
her legs without f'exing the hip.
Overbalancing is prevented by the close proximity of a table. or a ther
apist. and by well-timed instructions that help the individual make their
own segmental adjustment. For example, if observation of body align
ment suggests patient is losing balance backward the advice is to 'move
your hips forward'.
Computerized feedback training
A computerized footplate system with instrumented platform (e.g. Chattecx*)
provides the patient with feedback on position of COP as the possibilities of
moving in standing or sitting are explored (Fig. 2.18). An instrumented
platform allows the individual to practise moving the body mass voluntarily
with visual feedback and responding quickly to unexpected platform shifts.
This training may be useful in promoting a sense of control over body mass
and an understanding of the mechanical and functional effects of moving the
body about over the base of support. Responding to unexpected platform
perturbations may train individuals for moving walkways and escalators
(Horak et al. 1997).
"Chattecx Corporation, Hinton, TN.
68 Training gUidelines
FIGURE 2.17 (a) This reach involves hip, knee and ankle f'exion and extension and weight shift to
the L. (b) Moving a stool on wheels introduces the possibility of unexpected
movement. (c) At first she is doubtful whether she can pick up the cup but with
stand-by reassurance she can practise. Reaching with her R arm across the body moves
the body mass toward the L. (d) Practising with the cup on a box develops confidence to
reach even lower, (e) and to pick up objects from the floor, a difficult exercise for the
weak R limb. (f) He practises taking weight through his affected L leg. His R leg is also
weak from a previous injury to hip muscles so he practises to both sides.
(a) (b)

70 Training guidelines
FIGURE 2.18 Simple computerized devices can be used to give feedback about body position.
(Balance Performance Monitor, courtesy of SMS Technologies.)
<: :>
Patient may practise with minimal supervision.
A harness may be needed.
There is some evidence that computer-derived visual feedback related to
weight distribution is effective in gaining symmetrical stance following stroke.
However, there is only limited evidence that this training generalizes into
improved performance of functional actions (Winstein et al. 1989, Sackley
and Lincoln 1997). Given the specificity of exercise and training effects, it is
necessary for patients to practise utilizing the visual feedback natural to
different environmental conditions.
Maximizing skill
For balance to improve, individuals are challenged with increasingly difficult
tasks and more complex contexts in order to extend their abilities to the
limits. Flexibility in coping with a variety of real-life tasks and environmental
demands is introduced as soon as possible.
Reaching and pick-up exercises
place object beyond stability limit so it is necessary to take a step
increase weight of object
increase size of object to require two hands
include unpredictability
Balance 71
- vary speed of movement
- decrease the area of the base of support, e.g. one foot in front of
the other (Fig. 2.19a), standing on one leg (Fig. 2.19b), changing
compliance of support surface (see Siobounov and Newell 1993).
Stepping exercises
- standing, weight through non-paretic leg (repeat on affected leg)
stepping to markers on the floor (Fig. 2.19)
stepping on to high step
foot on ball, move ball about with affected leg.
Games to make rapid response time an imperative
- catching, throwing a ball, bouncing a ball, bat and ball games
involving having to step out.
Introduce complexity and uncertainty into the environment
- playing ball games in a group
- negotiating an obstacle course (see Fig. 3.27).
- stepping over obstacles of different sizes (Fig. 3.28).
Soft tissue stretching
It is critical that contracture and stiffness of calf muscles, particularly soleus,
are prevented as these problems restrict the ability to balance and may be a
cause of falls in elderly individuals. Short calf muscles interfere with hip
extension in standing and walking and a short soleus interferes with initial
foot placement backward in standing up. Exercises to prevent these
problems are as follows.
Sustained passive stretch to calf muscles in standing (Fig. 3.18a,c)
Sustained passive stretch to soleus muscle in sitting (Fig. 4.13)
Active stretch to calf muscles and hip flexors while exercising in standing
(Fig. 3.18d).
Strength training
Lower limb strength training is carried out individually, in a group and as part
of circuit training. The goal is for these exercises to transfer into improved
performance of actions performed in both sitting and standing and walking.
The exercises listed below are described in Chapter 3.
Step-up exercises (Fig. 3.22)
Heels raise and lower (Fig. 3.24)
Non-weightbearing exercises (Fig. 3.26).
Getting up from the floor
Teaching individuals how to get up from the floor after a fall is often a
priority in rehabilitation involving elderly individuals, although there is some
evidence that it is rarely done (Simpson and Salkin 1993). A major problem
is that many people after stroke have difficulty rising from the floor or find
72 Training guidelines
FIGURE 2.19 (a) One station in a circuit training class involves stepping forward with a narrow
base, and practising various tasks such as reaching to pick up an object. (Courtesy
FMackey, Physiotherapy Department, Port Kembla Hospital.) (b) Stepping on to a
plastic cup requires careful foot placement and continuous loading of the affected leg.
The exercise can also be done with the affected leg to train control. (c) Stepping
exercises should be started as soon as possible and practised to both sides. Stepping
across the other leg is specifically trained as it is a difficult action for people with poor
balance and can couse tripping. (Courtesy of K Schurr and S Dorsch, Physiotherapy
Deportment, Bankstown-Lidcombe Hospital, Sydney.)
(a) (b)
Balance 73
the action impossible unless they have good strength in the lower limb
extensors and arm muscles. There is now sufficient evidence to support the
need for intensive weight-resisted strengthening programmes for elderly
, leg. individuals with a history of falls or potential for falling, and in all patients with
motor impairments, in addition to balance training. Increasing muscle strength
increases the person's range of activities, assists in preventing falls and makes it
more likely that the person who falls is able to get up from the floor.
Measurement of balance provides objective information about an individual's
progress and is essential in clinical research to determine the effects of inter
vention. Tests used must be valid and reliable, and strict standardization fol
lowed. Choice of test, given that many are available, may be a dilemma for the
clinician which can be resolved by considering what information is required.
Choice of balance assessment is discussed by several authors (Maki and
McIlroy 1997, Carr and Shepherd 1998, Huxham et al. 2001).
Functional tests
The functional tests below measure the ability to balance the body mass during
self-initiated actions. They provide measures of different aspects of balance in
which the perturbations arise from actions performed in simple and stationary
environments. Most measure a single aspect of balance, while the Berg Scale
evaluates performance on 14 items common in daily life.
Functional Reach Test (Duncan et al. 1990, Weiner et al. 1993)
Step Test (Hill et al. 1996)
Timed 'Up and Go' Test (Podsialo and Richardson 1991)
Berg Balance Scale (Berg et al. 1989, 1992, Stevenson and Garland 1996)
Motor Assessment Scale: Sitting balance item (Carr et al. 1985)
Falls Efficacy Scale (Tinetti et al. 1993)
Activities-Specific Balance Confidence (ABC) Scale (Powell and Myers
1995, Myers et al. 1998)
Obstacle Course Test (Means et al. 1996)
There is no single test that can measure all aspects of balance, and in general
the tests listed above seek specific information. For example:
The Functional Reach test (FR) measures the distance that can be reached in
standing in a forward direction. It is therefore an indicator of the extent to
which the centre of body mass can be moved in a forward direction toward
the limit of stability.
The Timed 'Up and Go' test (TUG) measures the time taken to stand up
from a seat, walk off and sit down again. It is an indicator of the ability to
balance during body translation from one place to another, the ability to
balance being inferred by the time taken to perform the entire sequence. This
test is said to be a reliable predictor of the likelihood of falls (Shumway
Cook et al. 2000).
The Falls Efficacy scale measures the degree to which individuals fear
falling. The ABC scale measures a wider range of mobility-challenging
The fact that postural adjustments are an inherent part of an action has been
shown in many biomechanical studies. It is therefore important to understand
the specificity of the many tests, and not to expect the results to reflect
changes in aspects of balance not measured. One recent study (Wernick
Robinson et a1. 1999) showed no correlation between the score on the FR test
and gait velocity - a result which could be expected given that the two actions
are dynamically different.
Tests of walking are listed in Chapter 3. Walking tests do not specifically mea
sure balance but data about walking can provide inferences about balance. It
has been shown, for example, that stride time variability, measured with force
sensitive insoles, correlates with falling in elderly individuals, with the inference
__ Balance: clinical outcome studies
Reference Subjects Methods Duration Results
Lincoln 1997
Dean and
Weiss et al.
25 S5
6-31 weeks
Age range
41-85 yrs
20 Ss
>1 year
Age range
55-83 yrs
7 Ss
>1 year
Age >60yrs
Randomized controlled trial
E: Visual feedback training
on Nottingham Balance
Platform (NBP) (e.g. STS,
standing, reaching,
stepping) + task practice
(e.g. gait, climbing stairs)
C: Placebo programme on
NBP + task practice
Randomized controlled trial
E: Standardized training
programme of reaching
beyond arm's length in
C: Sham training of
cognitive-manipulative tasks
within arm's length
Time series
1: Baseline
2: Supervised high-intensity
progressive resistance
training for lower limb
4 weeks
12 sessions,
1 hour
2 weeks
10 sessions,
12 weeks
2 hours/week
At week 4, significant*
difference between groups on
stance symmetry, sway (NBP),
gross motor (Rivermead Motor
Scale), ADL (Nottingham 10-pt
ADL scale) *p < 0.05
Difference not retained at
12-week follow-up
E reached significantly* further
and faster, increased the load
through the affected leg in both
reaching and STS (force plate),
increased activation of leg
muscles (EMG) *p < 0.01. C did
not improve
Significant improvements in
muscle strength p < 0.01, time
taken on repetitive sit-to-stands
p < 0.02, Berg Balance Scale and
Motor Assessment Scale (gait)
p < 0.04, Medical Outcome
Survey (physical function)
p < 0.03. Significant correlations
between repeated sit-to-stands
and knee extensor strength
p < 0.04
ADL, activities of daily living; C, control group; COG, centre of gravity; E, experimental group; EMG, electromyography;
PT/OT, physiotherapy/occupational therapy; Ss, subjects; STS, sit-to-stand.
that this variable may be a measure of stability during walking (Hausdorff et al.
g 2001). The Obstacle Course, listed above, provides a complex environment for
measuring gait with inferences about walking balance.
Biomechanical tests
Biomechanical and muscle latency testing, including testing of postural responses
R test
to externally imposed perturbations and of the sensory contribution to balance,
is used in laboratory research to examine the effect of impairments on postural
adjustments. These tests are not typically used in the clinic. However, where an
instrumented force platform is available, the following tests may be useful in
guiding clinical interventions:
lce. It
force response to support surface movement (Chattecx Balance System*)
:rence effects of sensory inputs on balance (for the Clinical Test of Sensory
Interaction and Balance - CTSIBt)
Postural sway (force plate) (Note: functional relevance is unclear).
Evidence of effectiveness of some interventions is provided in Table 2.1.
Surprisingly few studies of the effects of interventions on balance after stroke
were found. Most investigations have tested the effect of weight-shifting prac
tice or sensory training on symmetry and postural sway. In general, patients
improved in the short term in the actions in which they were trained, with little
or no carry-over into improved function.
The most encouraging reports so far are from studies of the effects of task
oriented training of functional actions (Dean and Shepherd 1997, Dean et al.
2000) and of lower limb strength training (Sharp and Brouwer 1997, Weiss et al.
2000, Teixeira-Salmela et al. 1999, 2001), both of which have shown positive
effects on balance.
c: did
e and
*Chattecx Corporation, Hixton, TN.
tNeuro Com International, Inc, Clackamas, OR.
The ability to walk independently is a prerequisite for most daily activities. The
capacity to walk in a community setting requires the ability to walk at speeds
that enable an individual to cross the street in the time allotted by pedestrian
lights, to step on and off a moving walkway, in and out of automatic doors,
walk around furniture, under and over objects and negotiate kerbs. A walking
velocity of 1.1-1.5 rn/s is considered to be fast enough to function as a pedes
trian in different environmental and social contexts. It has been reported that
only 7% of patients discharged from rehabilitation met the criteria for commu
nity walking which included the ability to walk 500 m continuously at a speed
that would enable them to cross a road safely (Hill et al. 1997).
'delines for training
'king training
tissue stretching
'ting muscle activity
~ g t h training
'mizing skill

Walking is a complex, whole-body action that requires the cooperation of both
legs and coordination of a large number of muscles and joints to function
together. An important question in the study of motor control is how the many
structural and physiological elements or components cooperate to produce
coordinated walking in a changing physical and social environment. Bernstein
(1967) proposed that to perform a coordinated movement the nervous system
had to solve what he termed the 'degrees of freedom problem'. The degrees of
freedom of any system reflect the number of independent elements to be con
strained. Given the large number of joints and muscles and the mechanical com
plexity inherent in the multisegmental linkage, Bernstein questioned how these
elements can be organized to produce effective and efficient walking.
The coordination of a movement, Bernstein wrote, 'is a process of mastering
redundant degrees of freedom of the moving organism, that is, its conversion
to a controllable system'. Degrees of freedom can be reduced by linking muscles
and joints so that they act together as a single unit or synergy, thus simplifying
the coordination of movement. Such a linkage is illustrated in gait by the coop
eration between muscle forces at hip, knee and ankle to produce an overall
support moment of force to ensure the limb does not collapse.
Sensory inputs in general and visual inputs in particular provide information
that makes it possible for us to walk in varied, cluttered environments and
uneven terrains. Vision provides information almost instantaneously about
both static and dynamic features of the near and far environment. This
information is used to plan adjustments to the basic walking pattern (Patla
Avoidance of falls depends to a large extent on the ability to identify, predict and
act quickly enough upon potential threats to stability based on past experience.
The visual system plays an important role in identifying and avoiding potential
threats to balance. Avoidance strategies include adaptation of step length, width
and height, increasing ground clearance to avoid hitting an object on the ground,
increasing head clearance to avoid hitting an obstacle above ground, changing
direction and stopping (Patla 1997). These strategies are adaptive and are imple
mented to ensure stability of the moving body. Kinesthetic, tactile and vestibular
inputs also play an important role, particularly in the reactive control of balance
during walking. Reactive control by definition is a last resort or backup in
regaining balance and relies upon triggering of reflexive responses (Patla 1993,
Walking dysfunction is common in neurologically impaired individuals, arising
not only from the impairments associated with the lesion but also from sec
ondary cardiovascular and musculoskeletal consequences of disuse and phys
ical inactivity. Muscle weakness and paralysis, poor motor control and soft
tissue contracture are major contributors to walking dysfunction after stroke.
Functional gait performance, however, also depends on one's level of fitness.
Impairments post-stroke frequently impose excessive energy cost (effort) dur
ing walking, limiting the type and duration of activities. Stroke patients, particu
larly those of advanced age, are often unable to maintain their most efficient
gait speed comfortably for more than a very short distance, indicating that
muscle weakness, elevated energy demands and poor endurance further com
promise walking ability (Fisher and Gullikson 1978, Olney et al. 1986). Stroke
patients may self-select a speed that requires least energy (Grimby 1983) and
may not have the ability to increase this without increasing energy demands
beyond their capacity (Holden et al. 1986). Such individuals are constrained to
very limited activity in their home environment, and require the opportunity to
exercise and increase their walking speed and endurance.
Individuals who are discharged showing improvements in gait are not necessar
ily functional walkers. For example, the calculation of walking speed over 10 m,
a commonly used clinical measure of gait, may overestimate locomotor capacity
after stroke. Whereas healthy subjects can walk in excess of their comfortable
speed for at least 6 minutes, stroke subjects may not able to maintain their
comfortable walking speed over that time (Dean et al. 2001). This would prevent
them from being competent community walkers and may lead to increasing
handicap. Despite reports that 60-70% of stroke patients have regained inde
pendent walking function at rehabilitation discharge (Wade et al. 1987, Dean
and Mackey 1992), there is also evidence that only 15% report walking out
side the house 2 years later (Skilbeck et al. 1983).
During human gait the erect moving body is supported first by one leg then by
the other. As the moving body passes over the supporting leg, the other leg
swings forward for its next supportive phase. The gait cycle includes a stance
(approximately 60%) and a swing (approximately 40%) phase. The stance phase
can be further subdivided into weight acceptance, mid-stance and push-off, while
swing is divided into lift-off (early swing) and reach (late swing) (Winter 1987).
There are two brief periods of double support when both feet are in contact with
the ground and support of the body is transferred from the back leg to the for
ward leg. However, the body is supported on one leg for approximately 80% of
the gait cycle.
As the individual increases walking speed, the time spent in double support
decreases. Conversely, as walking speed slows, the duration of double support
increases. Double support phase is critical in transferring weight from one
limb to the other and in balancing the upper body. The time taken for
each step is similar, as is the distance covered with each step during forward
The balance requirements of gait are complex (see Ch. 2). In quiet standing the
projection from the body's centre of mass (COM), the centre of gravity
(COG), falls within the area of the feet. Throughout the gait cycle, however,
the path of the body's COG rarely passes within the perimeter of the foot
(Fig. 3.1). Apart from the brief double support period the body is in a potentially
unstable state. Safe foot clearance and foot contact are, therefore, essential to
rebalancing the body mass during double support (Winter et al. 1991). Figure
3.1 shows the path of the centre of pressure (COP), which is an indication of
weightbearing through the foot, during single limb support. It is notable that
the COP stays close to the centre of the foot and does not approach the outer
limits of the base of support.
The major requirements for successful walking (Forssberg 1982) are:
support of body mass by lower limbs
propulsion of the body in the intended direction
the production of a basic locomotor rhythm
dynamic balance control of the moving body
flexibility, i.e. the ability to adapt the movement to changing
environmental demands and goals.
FIGURE 3.1 A translational plot showing the body's COG and foot cOP for one walking stride.
~ v e n t
Note: the COG never passes over the foot. HC, heel contact; TO, toe-off.
(Reproduced from Winter et al. 1991, with permission.)
D by
r leg
Vc> of
; the
II to
~ r e
n of
The lower limbs play a significant role m these requirements. In the stance
phase they are involved in:
Support: the upper body is supported over one or both feet by the action
of lower limb extensor muscles and mechanical effects which prevent
lower limb collapse.
Propulsion (acceleration of the body in space): generation of mechanical
energy to drive the forward motion of the body in a basic locomotor
Balance: preservation of upright posture over the changing base of support
by postural adjustments occurring in lower limb and limb-trunk segmental
Absorption: mechanical energy is harnessed for shock absorption and to
decrease the body's forward velocity.
In the swing phase, in which the foot moves on a smooth path from toe-off to
heel contact, the lower limbs are involved in:
Toe clearance: to clear the foot from the ground.
Foot trajectory: to prepare the foot for a safe landing on the support surface.
These are the main functions that must be performed for effective gait (Winter
1987). Simplifying gait in this way not only identifies the major motor control
requirements but also clarifies for the clinician the critical requirements of
effective and efficient gait and the principal functions on which to concentrate
in training.
Gait has been studied more than any other everyday action, resulting in a very
large body of literature. Walking has been examined from many perspectives:
biomechanical, neural, pathological and developmental. Different types of gait
studied include stairwalking, negotiating kerbs and walking over obstacles.
Descriptive research efforts in gait laboratories have focused on quantifying per
formance in able-bodied subjects as well as examining common functionallimita
tions associated with different pathologies, with a view to developing appropriate
intervention strategies to be tested in the clinic. Since differences in such factors as
body mass, height, age, gender, cadence and context influence aspects of perfor
mance, large databases are required to provide 'normalized' values.
Spatiotemporal variables include cadence, stride length, step length and velocity.
When able-bodied individuals walk at their preferred speed, cadence varies from
a mean of 101 to 122 steps per minute and is reported to be slightly higher in
women than in men (Winter 1991). Stride length is approximately l.4m with a
step length of 0.7 m in able-bodied young subjects walking at their preferred
speed. Walking velocity for adult pedestrians aged between 20 and 60 years,
who were unaware they were being observed, was on average 1.4 m/s for males
and 1.2 m/s for females (Finley and Cody 1970).
Kinematic variables provide information about angular displacements, paths of
body parts and centre of body mass (CBM), with angular and linear velocities
and accelerations. These variables do not give information about the forces
that cause the movement. As Winter (1987) points out, the number of kin
ematic variables required to describe one gait stride is very high and some ser
ious compromise is necessary to make any gait description manageable.
In level walking, the path of the CBM describes a sinusoidal curve along the
plane of progression. The peak of the oscillations of this path occur at about
the middle of the stance phase. The path of the CBM also moves laterally and
describes a sinusoidal curve alternating from right to left in relation to support
of the weightbearing limb (Saunders et al. 1953). These motions are related to
each other in a systematic fashion, a smooth passage of the CBM being essen
tial for efficient walking.
Major angular displacements of lower limb joints during gait take place in the
sagittal plane as the body moves forward and can be seen by a careful observer.
80 100 60 40 20 o
2 0 ~
<5 0 4'.. ; .....~ ~ ..:::::::::::.
o n e .
-20 CV = 72% '., TO = 60%
Joint angle-natural cadence (N = 19)
Intersubject ensemble averages of hip, knee and ankle angles in the sagittal plane
during a single gait cycle. Flexion is positive, extension is negative. CV, coefficient of
variation; 0-60'Y0, stance phase; 60-100%, swing phase. (Reproduced from Winter
1987, University of Waterloo Press, with permission.)
In level walking, the pelvis rotates around a vertical axis with the magnitude
of this rotation approximately 4 on either side of the central axis. Since the
pelvis is itself a rigid structure, this rotation occurs alternately at each hip joint
which passes from relative internal rotation to external rotation in the stance
phase (Saunders et al. 1953). Rotation also occurs in the spinal joints. The
amount of rotation is related to stride length, with older individuals showing
less rotation than younger individuals (Murray 1967).
.. Hip
x 20 ....... CV=52%
ff. 0 > .
Sagittal plane angular displacements at hip, knee and ankle in able-bodied sub
jects walking at a natural cadence are shown in Figure 3.2. These average values
provide a set of reference norms that can be used to guide clinical observation
of gait. Normally individual variations in the magnitude of angular displace
ments are relatively small, with the greatest variations occurring in the less obvi
ous movements of ankle and foot (Sutherland et al. 1994).
The mass of the upper body, including the pelvis and trunk, responds to the
needs of the individual to advance the lower limb and to transfer body weight
from one supporting limb to the other. For example, the pelvis (moving at hip
joints and joints of lumbar spine) rotates, tilts, lists and shifts laterally, gov
erned by such factors as muscle length and length of stride. Shoulders rotate
and arms swing out of phase with displacement of pelvis and legs.
The pelvis also tilts forward and backward, movement occurring at hip joints
and joints of lumbar spine. Maximum forward tilting through a mean excur
sion of 3 occurs in mid to late stance and terminal swing. The movement is
controlled by gravity, inertia and hip flexor and extensor muscle action
(Sutherland et a1. 1994), as well as soft tissue length.
Lateral horizontal pelvic shift occurs as the CBM is displaced laterally in
preparation for single support. This displacement involves adduction at the
stance hip. The tibiofemoral angle acts as a mechanical constraint, allowing
the tibia to remain vertical while the hip adducts slightly. The extent of lateral
shift (4-5 cm) is governed in part by activation of hip abductors and muscles
extending the knee by mid-stance.
In association with this shift, the pelvis lists downward relative to the horizon
tal plane toward the side opposite to the stance limb. The displacement occurs
at the hip joints, with adduction of the stance hip and abduction of the swing
hip, and at joints of the lumbar spine. To overcome the effect of this 'leg
lengthening' on the swing leg, the knee joint of that leg flexes to 'shorten the
limb' and allow it to swing through (Saunders et al. 1953).
Trunk movement, in terms of specific movements of spinal joints during gait,
has not been studied extensively; however, current research utilizing three
dimensional measurement systems is enabling more discrete examination of the
complex spinal movements. The extent of trunk rotation about the longitudinal
axis may be greater than previously reported (Krebs et a1. 1992) and may be an
optimal means of increasing step length when conditions demand. A study in
which the movement patterns of lower thoracic, lumbar and pelvic segments
were examined found consistent patterns within and between different seg
ments. When walking speed increased, range of motion in each segment
increased. The results suggest that spinal movements are linked to motion of the
pelvis and lower limbs (Crosbie et a1. 1997a,b). This study also found a signifi
cant decrease in spinal range of motion in elderly individuals during walking.
These variables are calculated from force plate and kinematic data. They include
vertical and horizontal ground reaction forces (GRF) (Newtons per kilogram),
moments of force at the joints (Newton-metres per kilogram), power transferred
between body segments (watts), and mechanical energy of body segments
(joules). Moments of force are the net result of muscular, ligamentous and fric
tion forces acting to change the angular rotation at a joint. Net moments can be
interpreted as resulting principally from muscle forces (Winter 1991).
GRFs are the simplest kinetic variables to record. Although internally gener
ated forces are required to move body segments, the forces responsible for dis
placement of the body mass are external to the system. In an action such as
walking in which the feet are on the ground, these forces are called ground or
supportive forces. They occur in response to muscle activity and segmental
movements when body weight is transmitted through the feet (Fig. 3.3).
FIGURE 3.3 Averaged horizontal and vertical ground reaction forces. Horizontal force has a
negative phase during the first half of stance, indicating a slowing down of the body,
IS and a positive phase, indicating forward acceleration of the body, during the latter
half of stance. Vertical force has the characteristic double hump, the first related to
weight-acceptance and the second to push-off. CV, coefficient of variation.
(Reproduced from Winter 1987, University of Waterloo Press, with permission.)
~ e
~ e
~ e
~ e
Ground reaction forces-NAT. CAD. (N = 19)
TOE-OFF = 60%
20 40 60 80 100
% of stride
There are many combinations of muscle forces occurring across hip, knee and
ankle joints that can result in similar patterns of movement. There are also
many combinations of muscle activity that can produce the same moments of
force at a specific joint. However, in the stance phase, muscles across the three
joints act cooperatively and ensure support of the body mass through the
lower limb and prevent collapse of the limb. Winter (1980) has shown that
collapse is prevented by an overall extensor moment of force called a support
moment of force (SM). SM, an algebraic summation of forces at hip, knee and
ankle, is consistently extensor despite the variability that may occur at the indi
vidual joints. While ankle moment of force remains relatively consistent at dif
ferent speeds, hip and knee moments covary, a decrease in extensor moment
at the knee, for example, being countered by an increase in extensor moment at
the hip.
Mechanical power is the rate at which energy is generated and absorbed and is
the product of the net moment and angular velocity at the joint. An analysis of
power provides information about which muscles are generating mechanical
energy and which are absorbing it (Winter 1987). Power is generated when
muscle contraction is concentric (positive power) and is absorbed when muscle
contraction is eccentric (negative power). Power profiles of the three lower
Muscle activity
profiles of hip, knee and ankle during a single gait cycle indicating the most
important power phases. Power generation is positive, absorption is negative. CV,
coefficient of variation. (Reproduced from Winter 1987, University of Waterloo Press,
with permission.)
100 80
WM22 TOE-OFF = 60%
% of stride
20 o
o ..

limb joints in able-bodied subjects are shown in Figure 3.4. These profiles
describe the magnitude and direction of the flow of energy in and out of mus
cle and reveal that the major power generation occurring in gait occurs at the
ankle for push-off (about 80%), with the second most important source being
the generation of a flexion force at the hip for pull-off during late stance and
early swing (Winter 1983). These two major bursts of power are said to pro
vide the propulsive force necessary for forward level progression. The actual
energy to swing the leg through comes from pendulum action (the conversion
of potential energy to kinetic). Magnitudes of power profiles vary according to
speed but tend to remain constant in shape.
c 'l H3 Hip
... ....
o d'5iiko
. ......'
2 t', H2
Power Gen/ABS-natural cadence (n =9)
OJ (j) 0 '.y. .... .
<..'J I to :J""'i:.z;;;:pr K ):
. 7
Evidence suggests that, when subjects walk at their preferred speed, muscles
produce only a small amount of the total force required to propel the body
mass forward, with other contributions coming from ground reaction and iner
tial forces. Muscles contract and relax in a precise, orchestrated manner (Rab
1994). Their major role is to provide bursts of muscle activity at the appropri
ate time to initiate, accelerate, decelerate and control movement throughout the
gait cycle. As the cycle progresses, muscle action becomes primarily isometric or
eccentric, both of which are energy efficient forms of contraction to maintain
in EMG pattern of six major muscle groups associated with three different
gait cadences (n = 11). Note the change in EMG pattern is primarily in amplitude, the
shape remaining similar. The degree of amplitude change was muscle specific.
(Reproduced from Yang JF, Winter DA (1985) Surface fMG profiles during different
walking cadences in humans. Electroencephalography and Clinical
Neurophysiology, 60, 485-491, with permission.)
Across subject ensemble averages
Rectus femoris
Vastus lateralis

Lateral hamstring
e 200
,,-', -------CAD = 95
o e
o 00 00 100
% stride
upright posture against gravity and to transfer energy between segments.

Shortening (concentric) contractions are used to provide brief bursts of power
for forward motion when required (Winter 1987).
Surface electromyography (EMG) enables the determination, using a threshold
signal, of when in the gait cycle a given muscle or muscle group is active. Since
many factors affect the magnitude of EMG, including sensitivity to gait speed,
electrode size, shape, placement and amount of fatty tissue, there are numerous
problems in recording and reporting the electrical signals.
Winter (1991) reports EMG values for 25 lower limb muscles during gait. There
is considerable variability across subjects, some of which can be explained by
cles biological differences. Ankle muscle activity appears the most consistent, which
ody may reflect the joint's position closest to the base of support. Knee and hip muscles,
ler however, have multiple roles, particularly during stance, and tend to show more
variability. Signals from monoarticular muscles are less variable than those from
pn biarticular muscles, which may reflect different roles in producing power bursts
the (monoarticular muscles) and transferring power between segments (biarticular
: or muscles) (lngen Schenau et al. 1987). However, the patterns of EMG for
tain five major muscles at three different cadences (Fig. 3.5) indicate that the shape of
" Tibialis anterior
200 "',
'" .!,,# . ,
o ... - -_....... ....."'t. _
the signal remains similar despite changes in amplitude due to variations III
Walking in different environments
Slopes are frequently encountered in everyday life. Although muscle weakness
and many joint pathologies are known to interfere with slope walking, few stud
ies have investigated the effects of slopes on gait. Ramps are increasingly being
provided for access to buildings, one of the intentions being to reduce the num
ber of falls occurring on steps. Ramps, however, provide their own hazards. This
is particularly the case when walking down the slope due to the potential for
slippage and loss of balance. Slippage is a problem when walking down because
of the generation of high shear forces which increase as ramp angle increases.
The most significant finding in a study of urban pedestrians walking up and
down slopes of up to 9 was that step length decreased with increasing slope in
descent (Sun et al. 1996). This may reflect the need to accommodate the
increased friction demand with increasing slope. Another study comparing able
bodied subjects' level walking with downhill walking at a gradient of -19%
indicated that kinematic adjustments for downhill walking occurred primarily
at the knee joint (increased flexion) during stance and at the ankle and hip
joints during swing to compensate for the slope. Peak moments and powers
were much higher at the knee joints in downhill walking compared with level
walking (Kuster et al. 1995, Redfern and Di Pasquale 1997). Ankle dorsiflexion
moments also increase with slope angle during the first 20% of stance (Redfern
and Di Pasquale 1997). In walking uphill, subjects demonstrated a reduction in
walking speed, cadence and step length with increasing slope. In walking up a
slope with a 20% gradient, increases in hip and knee flexion were reported dur
ing late swing and the initial part of support phase (Wall et al. 1981).
Obstacles and kerbs
A common obstacle encountered in everyday life is a kerb or a single step
indoors. Negotiating obstacles has been investigated in the laboratory (Patla
et al. 1991, Grabiner et al. 1993). Obstacle crossing requires many adaptations
that depend on such factors as height, width and nature of obstacle, gait speed,
degree of lighting and position of object within step or stride. Modification to
gait may include changes in foot clearance, in step length and step time, and in
gait speed. Obstacle crossing requires clearance of an obstacle by both the lead
limb (clears the object first) and the trail limb.
A recent study observed a group of young able-bodied subjects negotiating a stan
dard kerb on the edge of a footpath in a natural setting outdoors (Crosbie and Ko
2000). All subjects were able to make accurate and appropriate adjustments to
their step pattern to negotiate the obstacle without any loss of fluency, regard
less of gait speed or cadence. Planning for obstacle negotiation appears to be
made some distance from the obstacle. Several critical requirements emerged
from analysis of the data. During kerb ascent, enough of the foot is placed on
Ip a
~ d ,
1 to
, be
top of the kerb to enable balance and stability during the swing phase of the
contralateral limb, and to enable propulsion from the stance foot once the kerb
had been negotiated. Another imperative appears to be that the toe of the swing
limb does not catch on the kerb as the foot is swung upward and forward. Kerb
descent appeared to require a more conservative step adjustment in which the
foot of the swing limb was either well short of, or well over, the kerb.
Stair walking
Stair walking is mechanically different from level walking in terms of joint
range, muscle activity and joint forces. Lowering and raising the body mass
while progressing forward to another step level requires increased angular dis
placements at the lower limb joints, increased duration and intensity of muscle
activity and increased magnitude of joint forces (Andriacchi et al. 1980).
Considerable demands are also placed on balance, particularly during the
period of single stance when the body mass is moving forward.
An essential feature of stair ascent is the forward translation of the body mass
over the foot by ankle dorsiflexion. The body remains in the same relative align
ment while forces produced by the back limb push the body mass diagonally in a
forward and upward direction over the forward foot. Extension at the hip, knee
and ankle of this leg raises the body mass as the back foot is lifted forward on to
the next step. Thus a smooth ascent results from the coordination of forces pro
duced by both legs. In contrast, during stair descent, the body mass is kept back
over the supporting leg which flexes at hip, knee and ankle to lower the body for
foot placement on to the next step. Stair descent has a shorter double support
phase than stair ascent, probably reflecting different mechanical demands.
Extensor muscle strength is particularly critical to stair walking since the body
mass must be raised or lowered essentially supported on one limb. In contrast
to level walking, it is the knee which generates most of the energy (Bradford
et al. 1988). During ascent, concentric contraction of lower limb extensor mus
cles raises the body mass vertically; stair descent is achieved through control of
gravitational forces by eccentric muscle contraction. Stair ascent, descent and
level walking all exhibit a support moment of similar shape. However, the
magnitude of support moment for stair walking is greater. During ascent it is
twice the magnitude produced during level walking (Bradford et al. 1988).
The ability to maintain effective gait function with increasing age is important
for the health and well-being of older people. Although it is a common obser
vation that the walking pattern changes with increasing age, how many of the
changes observed in the older individual are due to an ageing process and how
many to reduced physical activity and the pathology that may accompany age
ing is not clearly understood. Due to fear of falling and the possible conse
quences of a fall, apprehension may also influence performance. The changes
observed in walking pattern may also reflect slower walking speeds commonly
reported in the elderly. Cross-sectional studies have repeatedly shown decreases
in walking speed and stride length, and increases in duration of double support
phase in older subjects.
Investigators have also described age-related decreases in vision, cardiovascular
capacity, muscle strength, joint flexibility and bone mass, all of which could
impact on walking performance. Many of the above decrements are similar to
those associated with extended periods of bed rest and relative inactivity (Bortz
1982). Gabell and Nayak (1984), using stringent subject selection criteria, failed
to find significant age effects on spatiotemporal gait variables in a group of
healthy elderly subjects aged between 66 and 84, suggesting that pathology
rather than age may be the major contributor to variability in gait. Stride width
and duration of double support phase, however, showed less consistency than
step length and stride time.
Falls are a frequent hazard for elderly individuals. The reduction in frequency of
falls among the elderly is a public health issue and a major goal of researchers
and health workers. Epidemiological data have implicated initiation of walking,
turning, walking on uneven surfaces and stopping in almost all incidences of
falls (Prudham and Evans 1981, Gabell et al. 1986). Increased stride width,
erratic floor clearance and increased stride time variability (Hausdorff et al.
2001) are three parameters that distinguish between fallers and non-fallers.
Recovery from an obstacle-induced anteriorly directed stumble appears to
depend upon lower extremity muscle power and ability to restore control of
hipltrunk flexion (Grabiner et al. 1993). Individuals with limited ankle joint
mobility may be particularly at risk of stumbling, tripping and falling. Range of
motion and joint stiffness studies suggest that decreased mobility at the ankle
joint may be a characteristic of ageing associated with an increasingly sedentary
life style, and acute and chronic pathology (Vandervoort et al. 1990). Muscle
weakness, especially in the ankle musculature, seems to be another factor con
tributing to falls and variability in gait. Reduced push-off power in subjects
aged between 62 and 78 years compared with younger subjects has been
reported (Winter et al. 1990), and significantly smaller peak torque and power
values were found in ankle and knee muscles in a group of fallers compared
with a group of non-fallers (Whipple et al. 1987).
Several studies of older individuals show beneficial effects of muscle strengthen
ing exercises and aerobic exercises in preventing falls, increasing gait speed, and
aerobic capacity (Fiataroni et al. 1990, Krebs et al. 1998). Vandervoort (1999)
reported positive effects of strengthening and stretching exercises even in very old
residents in long-term care. Interestingly, in a longitudinal study of aerobic capac
ity of master athletes aged between 50 and 82 years, those who continued com
petitive levels of training maintained their aerobic capacity. Those who decreased
training showed significant decreases in aerobic capacity (Pollock et al. 1987).
Certain gait abnormalities are evident following stroke. In part they arise as a
result of different patterns and degrees of muscle weakness/paralysis and lack of
Jort coordination, and in part as a result of adaptations as the person attempts to walk
despite muscle weakness, balance deficits, changes in muscle length and stiffness.
Over the past decade or so the increasing use of movement analysis systems
that enable measurement of kinematic and kinetic variables, together with
EMG analysis, have provided data that not only have greatly increased our
understanding of gait adaptations associated with neurological dysfunction but
have also enabled clinical gait analysis to be more objective. Descriptive vari
) of
ables for measurement include gait speed, stride length, cadence and joint
angles, whereas moments of force and power profiles along with EMG signals
help to explain why a given movement pattern might occur.
Research findings
~ e r s
In an attempt to investigate whether or not there is a typical pattern of muscle
weakness in the lower limbs, Adams et al. (1990), using myometry, examined
s of
the voluntary strength of eight muscles of the lower limb (flexors and extensors
of hip, knee, ankle and hallux) in able-bodied subjects and in both the weak
: al.
and clinically unaffected lower limbs of stroke subjects. No consistent pattern
of muscle weakness was found in the hemiplegic limb. Sometimes the flexors
were more affected than the extensors at one joint with the converse at another
, to
joint. The strength of muscles tested on the clinically unaffected side was also
I of
reduced compared with controls. On average, one of the most affected groups
was the ankle plantarflexors.
e of
Given the high energy cost of walking in the presence of muscle weakness,
decreased motor control and adaptive soft tissue changes it is not surprising
that the most consistent research finding is that individuals following stroke
walk more slowly than able-bodied age-matched subjects (e.g. von Schroeder
et al. 1995). Slow walking disrupts the dynamics of the action and many of these
spatiotemporal characteristics are also found in able-bodied subjects walking at
these very slow speeds. Consistent with a decrease in speed are decreases in
stride length and changes in cadence, and increased duration of double support
phase (Olney and Richards 1996). Several differences in proportion of time
spent in stance and swing phases have been reported. A typical finding is that
the stance phase of the non-paretic limb is both longer and occupies a greater
proportion of the gait cycle than that of the paretic limb.
Studies of joint angle displacements following stroke, focused in general on
sagittal plane movements, have shown variability in magnitude of angular dis
placements at a particular point in the gait cycle compared with able-bodied
subjects. The major differences are (Burdett et al. 1988, Lehman et al. 1987):
decreased hip extension at the end of stance
decreased hip flexion in mid-swing
decreased knee flexion at toe-off and mid-swing
decreased ankle dorsiflexion at foot contact, and during stance associated
with a hyperextended knee
lS a
decreased ankle plantarflexion at toe-off
( of
increased knee flexion at foot contact.
Note that some of these changes in angular displacement can be attributed to
differences in walking speed. The data from one group of stroke subjects were
divided into three groups according to speed. In the fastest group, the hip
extended an average of 10 beyond neutral in late stance, but only about r
beyond neutral in the medium speed group. Extension was a few degrees
short of neutral in the slowest group (Olney and Richards 1996). The clinically
unaffected limb showed similar trends although the average hip extension
always exceeded neutral.
While spatiotemporal and kinematic variables have descriptive value in clinical
practice, kinetic variables have particular explanatory value for our understand
ing of the underlying reasons for a particular movement pattern. Kinetic vari
ables are of value in determining, for an individual patient, where in the support
phase to focus intervention. They can also be used to evaluate the effect of
Several vertical GRF variations in stroke patients have been reported compared
with able-bodied subjects (Carlsoo et al. 1974). These variations included
decreased magnitude of vertical GRFs, particularly associated with push-off.
Rather than the characteristic two peaks and intermediate trough there was a
relatively constant magnitude of force throughout the support phase. As an
example of how such information is of clinical value, the absence of vertical
GRF peak representing push-off suggests the need to strengthen the calf muscles,
increase their length, and emphasize push-off in walking training.
The propulsive force for forward progression comes mainly from generation of
power at the ankle for push-off and at the hip for pull-off (Winter 1987). A con
sistent finding in able-bodied individuals is a relationship between gait speed and
power bursts at pull-off and toe-off. A recent study provides profiles of moments
of force and powers at hip, knee and ankle for two hemiplegic subjects compared
with values obtained in slow walking in able-bodied subjects (Richards et al.
1999). One patient who walked very slowly (0.15 m/s), maintained hip, knee and
ankle flexion throughout most of the gait cycle. This was associated with high
extensor moments at hip and knee and low or absent plantarflexor moments.
Power generation was very low at push-off and pull-off. In contrast, profiles of a
second hemiplegic subject walking a little faster (0.57 m/s) showed mechanical
power values that were much closer to slow normal values but less sustained
than normal. It is noteworthy for clinical practice that significant increases in
push-off and pull-off power generation and in walking speed have been teported
after 4 weeks of intensive task-specific exercise and training which included muscle
strength training and treadmill walking (Dean et al. 2000).
Olney and Richards (1996) also report exaggerated extensor moments at hip
and knee throughout stance. It is often observable in the clinic that the hemiplegic
leg is held stiffly during stance. This stiffness may be misinterpreted as reflecting
'spasticity'. However, the exaggerated extensor moments may reflect increasing
muscle co-contraction to prevent collapse of the limb. This would indicate the
need for weight bearing exercises to strengthen lower limb muscles and improve
control of the limb during both concentric and eccentric activity (i.e. flexion and
extension) in addition to walking practice.
I to There is a great deal of variability in the reporting of EMG activity in stroke
ere patients, reflecting differences in speed, motor control deficits and adaptations
hip (Peat et al. 1976, Knutsson and Richards 1979). Low levels of muscle activity
2 have been reported in some muscles (e.g. gastrocnemius). In other muscles, levels
ees of activity were increased in some phases of the gait cycle and decreased in oth
lily ers. As would be expected, abnormal patterns of EMG activity during walking
Ion have also been reported for the clinically unaffected limb (e.g. Shiavi 1985).
cal Observational analysis
In A clear distinction should be made between measurement, i.e. the collection of
ort objective data, and observational analysis. Visual observation, which therapists
of must rely on for ongoing gait analysis and as a guide to intervention, is depend
ent on the clinician's skill and knowledge of normative biomechanical values.
There are some major drawbacks to observational assessment. Critical bio
red mechanical characteristics such as angular velocities, ground reaction and muscle
led forces cannot be observed but they can be inferred. Several clinical studies have
>ff. indicated the problems, showing lack of reliability in observational assessment
5 a (Wall 1999), erroneous and incorrect conclusions (Malouin 1995), and lack of
an familiarity with normative values (Eastlake et al. 1991).
!es, It is not possible or practical for every individual with stroke to have a full bio
mechanical gait assessment. However, researchers are continuing to investigate
the nature of gait deficits and attempting to explain the reasons for functional
of limitations observed by clinicians. This ongoing research continues to direct
clinical practice.
Ilts In the clinic, the only directly observable movement characteristics are kinematic,
'ed particularly angular displacements and paths of body parts, and such spatio
al. temporal characteristics as speed, step length and stride width. Spatiotemporal
nd and kinematic variables can also give clues to the underlying dynamics. For
gh example, collapse of the lower limb at the knee suggests muscle weakness and
ts. insufficient control of extensor forces. In order to increase observational accu
fa racy, both sagittal and frontal views are necessary for the clinician to analyse
::aI the relationships between multiple segments. Videotaping is recommended to
ed enable multiple viewing opportunities and avoid the patient becoming frus
trated and fatigued.
:le Biomechanical studies indicate that there are several key kinematic characteristics
that are necessary for effective and efficient gait. These characteristics or compo
nents are based on data provided by able-bodied subjects walking at their pre
ferred speed. Although there are variations between individuals or within the one
individual as a result of such factors as change in speed or alteration in footwear,
there are certain observable events shared by all (Fig. 3.6). Individual variations,
for example in the magnitude of angular displacements are relatively small.
ve The essential components listed below provide a basic structure or model
Id upon which observational analysis is carried out. The patient's attempts at
walking (the performance) are observed and compared to this 'normal' model.
92 Training gUidelines
FIGURE 3.6 Gait cycle in an able-bodied subject illustrating major kinematic components in the
sagittal plane.
Although components are listed separately, they are of course integrated during
the gait cycle.
Stance phase
Ankle dorsiflexion for heel contact, plantarflexion to foot flat, dorsiflexion
as body mass progresses forward over the foot, pre-swing ankle
plantarflexion for push-off
Knee flexion at heel contact (approximately 15) to absorb body weight
and momentum, extension by mid-stance, flexion (35-40 pre-swing)
Hip extension to advance body mass forward over the foot (approximately
10-15 beyond neutral) together with ankle dorsiflexion
Lateral horizontal pelvic shift (approximately 4 em from side to side)
involving adduction at stance hip
Transition from stance to swing
Several events in late stance appear to set up the conditions for swing. Hip
extension in late stance is associated with moving the body mass forward over
the stance foot, which provides the hip flexors with a mechanical advantage to
generate power for pull-off at the start of swing (Teixeira-Salmela et al. 2001).
Step length and walking velocity depend principally on plantarflexor activity at
push-off and hip flexor activity at pull-off (Winter 1987).
Swing phase
Hip flexion (pull-off) to swing lower limb forward
Knee flexion to shorten lower limb (35-40 pre-swing, increasing to 60)
Lateral pelvic list downward (approximately 5) toward swing side at toe-off
Rotation of pelvis about the vertical axis (approximately 4 to either side)
Knee extension plus ankle dorsiflexion for heel contact
Walking 93
FIGURE 3.7 Normally the knee flexes at the start of stance and provides shock absorption. Note
the hyperextension of the knee. (Reproduced from Perry J(1992) Gait Analysis.
Normal and Pathological Function. Slack, Thorofare, NJ, with permission.)
The challenges to stability during locomotion are complex. They present major
difficulties for the patient and can instill fear of walking. The base of support
during both single and double support phases is small, and the CBM is a con
,Iv siderable distance above it. Balance control is concentrated in muscles linking
thigh to pelvis and shank to foot.
The individual's ability to walk varies with the severity of impairments, the
ability to utilize adaptive movements and the extent of secondary soft tissue
contracture. Although the mechanisms underlying gait limitations are complex
and multifactorial, it is possible to identify some causes of observed functional
limitations to guide intervention (see also Adams and Perry 1994, Olney and
Richards 1996).
Typical kinematic deviations and adaptations
Initial stance (heel/foot contact and loading)
Limited ankle dorsiflexion
- decreased activation of anterior tibial muscles
- contracture and/or stiffness of calf muscles with premature activation.
Lack of knee flexion (knee hyperextension) (Fig. 3.7)
contracture of soleus
- limited control of quadriceps 0-15.
94 T raining gUidelines
FIGURE 3.8 (a) The knee remains flexed throughout stance. Soleus weakness fails to stabilize the
tibia. As a result, without a stable base the quadriceps cannot extend the flexed knee.
(b) The mechanism is made clear by the stick figure. (Reproduced from Perry j (1992)
Gait Analysis. Normal and Pathological Function. Slack, Thorofare, Nj, with
(a) (b)
Lack of knee extension (knee remains flexed 10-15
with excessive ankle
dorsiflexion) (Fig. 3.8)
decreased activation of calf muscles to control movement of shank
forward at ankle (ankle dorsiflexion)
limited synergic activation of lower limb extensor muscles.
Stiffening of knee (hyperextension). This interferes with preparation for
contracture of soleus (Fig. 3.9)
- an adaptation to fear of limb collapse due to weakness of muscles
controlling the knee.
Limited hip extension and ankle dorsiflexion with failure to progress body
mass forward over the foot
- contracture of soleus (Figs 3.9, 3.10).
Excessive lateral pelvic shift (Fig. 3.11)
decreased ability to activate stance hip abductors and control hip and
knee extensors.
Walking 95
FIGURE 3.9 Lack of forward movement of the shank on the foot (ankle dorsiflexion) and knee
hyperextension are compensated for by forward movement of the trunk at the hip
(hip flexion). Note the short contralateral step. (Reproduced from Perry j (1992) Gait
Analysis. Normal and Pathological Function. Slack, Thorofare, Nj, with permission.)
FIGURE 3.10 Left mid-stance. Failure to transport body mass over the foot due to a short soleus
which prevents ankle dorsiflexion and hip extension and hyperextends the knee.
96 Training gUidelines
FIGURE 3.11 Increased lateral pelvic shift to the R and downward list of the pelvis on the L.
FIGURE 3.12 Bold limb shows inadequate knee f'exion, and ankle plantarflexion with prolonged
foot contact in pre-swing. (Reproduced from Perry J (1992) Gait Analysis.
Normal and Pathological Function. Slack, Thorofare, NJ, with permission.)
Walking 97
FIGURE 3.13 (a,b) Inadequate hip and knee f'exion at the beginning of swing with toe drag (bold
limb). (Stick figure from Perry J(1992) Gait Analysis. Normal and Pathological
Function. Slack, Thorofare, NJ, with permission.)
(a) (b)
Late stance (pre-swing)
Lack of knee flexion and ankle plantarflexion, prerequisites for push-off
and preparation for swing (Fig. 3.12)
- weakness of calf muscles.
Early and mid-swing
Limited knee flexion (Fig. 3.13). Normally 35-40 increasing to 60 for
swing and toe clearance
increased stiffness in or unopposed activity of two-joint rectus femoris
- decreased activation of hamstrings.
The mechanisms that prevent knee flexion in pre-swing reduce the potential for
normal limb advancement. Insufficient knee flexion requires adaptive action of
98 Training guidelines
FIGURE 3.14 Backward trunk movement with posterior pelvic tilt to advance the leg forward when
hip flexion is reduced. (Reproduced from Perry J(1992) Gait Analysis. Normal and
Pathological Function. Slack, Thorofare, NJ, with permission.)
adjacent segments to achieve limb advancement and foot clearance (Adams
and Perry 1994). Typical adaptations to limited knee and hip flexion are hip
hiking and increased posterior pelvic tilt, or limb circumduction.
Limited hip flexion at toe-off and mid-swing (Fig. 3.14)
- decreased activation of hip flexors.
Limited ankle dorsiflexion'"
may be slowness in flexing knee rather than decreased dorsiflexor
stiffness and contracture of calf.
Late swing (preparation for heel contact and loading)
Limited knee extension and ankle dorsiflexion jeopardizing heel contact
and weight-acceptance (Fig. 3.15)
contracted or stiff calf muscles
- decreased dorsiflexor activity.
". Strength requirements of dorslflexors vary throughout the gait cycle. Functional demands during
loading far exceed requirements during swing (Adams and Perry 1994).
Walking 99
FIGURE 3.15 (a) Normally the hip flexes, the knee extends and foot dorsiflexes as the leg reaches
out to make heel contact. Note the lack of ankle dorsif'exion and the very short step.
(b) The seven figures to the R illustrate one gait cycle.
FIGURE 3.16 A frontal plane view showing increased stride width. Note she is not moving her
body weight forward and diagonally to the L but keeping her weight back on her
non-paretic leg as she steps forward.
Training guidellinEls
Spatiotemporal adaptations
These include:
decreased walking speed
short and/or uneven step and stride lengths
increased stride width (Fig. 3.16)
increased double support phase
dependence on support through the hands.
Although symmetry is frequently an unquestioned goal of gait rehabilitation,
an investigation of 34 gait variables failed to provide evidence that symmetry
promotes an increase in walking speed (Griffin et al. 1995). It was pointed out
that there are several reasons for expecting to see asymmetric rather than sym
metric variables in individuals whose limbs have unequal capabilities. Knowing
gait is asymmetrical is, therefore, not helpful in planning intervention. The
goals should be directed toward impairments causing the asymmetry - weak
ness and lack of control of segmental movement in the context of walking.
Promoting the rhythmic nature of walking (e.g. on a treadmill) should be more
of a focus than symmetry which will improve when lower limb muscles are
stronger and gait speed is increased.
Intervention aims to optimize walking performance by:
preventing adaptive changes in lower limb soft tissues
eliciting voluntary activation in key muscle groups in lower limbs
increasing muscle strength and coordination
increasing walking velocity and endurance
maximizing skill, Le. increasing flexibility
increasing cardiovascular fitness.
These aims are addressed by a combination of treadmill and overground
walking practice, strength training and soft tissue stretching. In general, it is
the tendency for the stance limb to collapse, lack of forward propulsion and
poor balance that interfere most with gait. Although in the acute phase
walking capacity may be limited, it is only by integrating strength gains into
meaningful function (Le. walking) that performance will improve.
Given the mechanics of the action, the major emphasis in walking training
is on:
support of the body mass over the lower limbs
propulsion of the body mass
balance of the body mass as it progresses over one or both lower
controlling knee and toe paths for toe clearance and foot placement
optimizing rhythm and coordination.
Since support (loading the limb) is a prerequisite for upright stance, and
push-off sets up the conditions for propulsion and swing, training focuses
initially on these. Note that support, balance and stepping are trained only
when the individual practises walking itself. A treadmill with harness
provides some body weight support and safety, enabling early walking for
those who would otherwise not be able to practise. The use of hands for
balance and support is discouraged to enable the regaining of dynamic
balance and the ability to support the body mass through the lower limbs.
The following guidelines provide what can be considered a critical pathway
for optimizing gait for individuals following stroke. The guidelines are
individualized according to each patient's impairments, level of functional
ability and needs. The guidelines are set down with no intention of a
prescribed sequence. They address the essential requirements for effective
gait, synthesized from research. Many of the exercises can be practised as
part of circuit training or in a group with minimal supervision.
Walking training
When to commence walking practice and what criteria to use have been
critical clinical questions in stroke rehabilitation. Conventional gait training
has focused on part-practice of components of gait in preparation for
walking. Gait training on a treadmill with a harness and partial body weight
support, however, enables walking training to commence far earlier than has
been possible. It is also a method of enabling repetitive leg exercise. Patients
as early as 1 month after stroke have commenced training on a treadmill
(Malouin et al. 1992). Reservations have been expressed about the
studies so far, for example, the variations in methodology and small sample
sizes which may bias the size of the treatment effect (Dobkin 1999).
Without a doubt randomized controlled studies are necessary. However,
all of the studies show effective results, some showing substantially
better results than the usual methods of therapy provided in rehabilitation
(see Table 3.1).
Walking on a treadmill with harness and partial
body weight support (Fig. 3.17)
Adjust treadmill speed to achieve an optimal step length (Moseley and
Dean 2000). Treadmill speeds as low as 0.5 km/h or 0.1 m/s may be
required for some patients. However, if speed is too slow, a rhythmical,
cyclic gait pattern is not stimulated.
Use the minimum amount of body weight support and arm support to
maintain hip and knee extension of stance limb. A maximum of <30%
body weight support appears to result in the most normal gait pattern.
Any more than 50% body weight support results in the patient walking
on their toes (Hesse et al. 1997).
Decrease body weight support when patient is able to swing leg
__ Gait: clinical outcome studies
Reference Subjects Methods Duration Results
Hesse et al.
Sharp and
Brouwer 1997
Visintin et al.
Duncan et al.
7 Ss
> 3 months post-stroke
Mean age 60.3
Age range 52-72 yrs
15 Ss
> 9 months post-stroke
Mean age 67 10 yrs
79 Ss
(E: 43, C: 36)
< 6 months post-stroke
Mean age 65.2 11.1 yrs
20 Ss
30-90 days post-stroke
Mean age
C: 67.8 7.2 yrs
E: 67.5 9.6 yrs
A-B-A design
A: PWS treadmill training
B: Bobath
Pre-test, post-test design
Warm up, stretches, isokinetic
strengthening exercises for
paretic leg, cool down
Randomized controlled trial
Treadmill training:
E: PWS up to 40%
C: no PWS
Randomized controlled trial
E: Exercise programme (warm-up,
stretches, assisted/resisted exercises,
balance, progressive walking or
bicycle ergometer, functional
activities with upper limb)
C: Usual care (highly variable in
intensity, frequency and duration.
No subject received endurance
3 weeks each
A - 30 min/day
B - 45 min/day
6 weeks
3 days/week, 40 mins
6 weeks
On average 15 min/day
12 weeks (8 of which
were therapist
3 days/week, 1.5 hours
Gait velocity increased significantly during
treatment A only
Significant increases in muscle strength
(p < 0.05), gait velocity (p < 0.05), overall
physical activity scores (HAP) (p < 0.003)
Improvements in stair climbing and Timed
'Up and Go' test were not significant
No increase in spasticity
Follow-up at 4 weeks - gait velocity and HAP
scores continued to increase
79% of E Ss were training without PWS at end
of study
E scored significantly higher than C on
Berg Balance Scale (p = 0.001), overground
gait velocity (p = 0.02), overground walking
endurance (p = 0.01)
Follow-up at 3 months: E group continued to
have significantly higher scores for overground
gait velocity (p = 0.006)
Significant differences in gait speed*, Fugl-Meyer
scores for lower extremity* between E and C.
Changes in Berg score, 6 min walk test
favoured E but were not significant. No
between-group differences in Barthel Index,
Jebsen Test of Hand Function, Lawton ADL
Scale, MOS-36, apart from Physical Function
___ _
Reference Subjects Methods Duration Results
et al. 1999
Mercier et al.
Dean et al. 2000
et al. 2001
13 Ss
9 months post-stroke
Mean age
69.42 ::':: 8.85 yrs
1 S
19 months post-stroke
39 yrs
9 Ss
> 6 months post-stroke
Mean age
E: 68.8 ::':: 4.7 yrs
c: 64.8 ::':: 3.3 yrs
13 Ss
> 9 months post-stroke
Mean age 67.7 ::':: 9.2 yrs
Randomized controlled pre-test,
post-test design
Warm up, aerobic exercises, lower
limb strengthening, cool down
Single case design
Training on static dynamometer
(S produced submaximal static
forces in 16 sagittal plane directions
with computer-generated feedback
on direction and intensity of applied
Randomized pre-test, post-test
group design
E: Circuit training (reaching in sitting
and standing, sit-to-stand, step-ups,
heel lifts, isokinetic strengthening,
treadmill walking, walking over
obstacles, up and down slopes)
C: Training of affected upper limb
Pre-test, post-test design
Warm-up, aerobic exercises, lower
extremity muscle strengthening,
cool down
10 weeks
3 days/week
6 weeks
3 days/week, 1 hour
4 weeks
3 days/week, 1 hour
10 weeks
3 days/week
28.3% improvement in gait speed (p = 0.00);
37.4% improvement in stair climbing (stairs/min)
(p = 0.00); 42.3% improvement in mean peak
torque generated by all muscle groups of
affected leg (p = 0.00); 39.2% improvement in
functional activities on adjusted activity score
(HAP) (p < 0.001) No increase in spasticity
Improvements in Timed 'Up-and-Go' test
(26.9%), comfortable gait speed (72.0%),
2 min walk test (121.4%)
Follow-up at 6 weeks - some loss of function
which was still considerably improved
compared with baseline values
E scored significantly higher than C on gait
velocity and endurance, VGRF produced
through affected leg in STS and number of
repetitions in step test (p os: 0.05)
Follow up at 2 months - improvements retained
Gait speed increased 37.2%, associated with
increases in cadence and stride length
(p < 0.001)
Post-training profiles showed that subjects
generated higher power levels, with increases in
positive work by ankle plantarflexors, hip
flexor/extensor muscles
ADL, activities of daily living; C. control group; COG. centre of gravity; E. experimental group; HAP, Human Activity Profile; MOS-36, Medical Outcome Study-36 Health Status
Measurement; PWS. partial body-weight support; Ss, subjects; STS, sit-to-stand; VGRF. vertical ground reaction force.
104 Training guidelines
FIGURE 3.17 (a,b) Treadmill training with supportive harness. (b) The physiotherapist assists with
toe clearance and step length during swing phase of the paretic leg.
(a) (b)
Encourage patient to decrease arm support.
Increase treadmill speed as soon as possible, using clinical judgement,
ensuring that step length is not compromised, i.e. avoid an increase in
cadence with a decrease in step length.
Therapist may need to swing affected lower limb forward (Fig. 3.17b) and to
stabilize the knee in stance to avoid hyperextension.
Walking on a treadmill with harness and no
body weight support
Increase speed, slope and duration (with warm-up and cool-down periods).
Encourage decrease in arm support.
Walking 105
Encourage increase in step length.
Progress to treadmill walking without harness.
This training, if done intensively, can increase endurance and aerobic
A recent study demonstrated that increasing treadmill speed in a structured
manner (using an interval paradigm) led to significantly greater overground
walking speed than training with limited progression (Pohl et al. 2002).
A useful description of the advantages and disadvantages of training walking
on two commercially available machines (TR Spacetrainer* and Lite Gait
has recently been published by clinicians experienced in their use
(Crompton et al. 2001). Both the TR Spacetrainer and Lite-Gait can pull the
patient up from sitting to standing. This is a useful feature for heavy,
dependent patients and enables walking practice while complying with a no-lift
policy. The Lite-Gait has the additional capacity of use without a treadmill
for walking overground. The authors suggest that the harnesses provided
need to be more adjustable and available in different sizes.
Overground walking
A Lite-Gait may help the individual make the transition from treadmill
walking to walking overground. In particular, such an aid enables the
individual to focus on walking without fear of falling. The Lite-Gait frame
needs to be pushed by the therapist or aide since it is quite large and
difficult to manoeuvre.
Walking forward Starting with body erect, i.e. hips extended, patient steps
forward with non-affected limb then affected limb.
Initiating gait with affected limb in stance may set up the best conditions
for the subsequent swing phase.
The goal of the first few steps is for the patient to get the idea of pro
ducing a basic locomotor rhythm, while supporting, propelling and bal
ancing the body mass.
Steady patient at upper arm or at safety belt. Do not interfere with pro
gression forward.
Stand to the side of patient or if necessary behind, so as not to impede
Encourage patient to step out and take even steps. Foot prints on
the floor may give the idea of an optimal step length and stride width
and provide a stimulus to increasing step length and decreasing stride
"-TR Equipment, 5-57322 Trenas, Sweden.
I,v!obility Research, Tempe, Arizona, USA.
106 Training guidelines
Do not hold patient too firmly as this may offer resistance to translation
of body mass forward or interfere with dynamic balance of the linked
segments over the feet.
Do not confuse patient - give only brief instructions during walking; for
example, 'Take a longer step', 'Step out' or 'Just do it' may be enough to
stimulate an attempt at a person's 'normal' walking.
Encourage patient to walk faster and take longer steps. This may be
counter-intuitive to therapists who have worked with patients at
relatively slow speeds with the intention of improving the quality of
movement. It is difficult to develop an effective locomotor rhythm when
walking very slowly. Slow walking disrupts the dynamics of the action in
both able-bodied and disabled subjects and it has been demonstrated
that when stroke patients walk at their maximum speed they show
more symmetry (Wagenaar 1990). A significant relationship has been
found between speed and gait parameters (Wagenaar and Beek 1992).
An early goal for the patient can be to walk to the next appointment
as an alternative to being transported in a wheelchair, or at least part of
the way, with the explicit aim of increasing the distance walked each day.
Patient selects a distance to walk independently and safely. It may only
be the length of the bed initially. The goal is to walk this distance five
times a day, increasing it as soon as possible. The number of sessions
per day can be decreased as walking improves and distance increases. At
home, the patient continues with a regime of increasing both distance
walked and duration of walk.
As soon as the patient achieves a degree of independence, practice of
negotiating ramps, stairs, kerbs and obstacles needs to be introduced. The
evidence from follow-up studies is that few patients when discharged are
able or confident enough to walk ouside the house (Hill et al. 1997).
Patients are discharged from a flat, protected environment one day and
expected to deal with the complexities of the built and natural
environments the next day, often with little practice in such environments.
Soft tissue stretching
Preservation of the extensibility of soft tissue, in particular calf muscles and
rectus femoris, is critical to the patient's ability to stand up, walk and stair
walk. Preserving functional muscle length involves both passive and active
Brief sustained passive stretches are carried out immediately before exercise
to decrease muscle stiffness. Active stretching occurs during the
strengthening exercises given below.
Calf muscles: passive stretch in standing (Fig. 3.180)
Rectus femoris: passive stretch in prone (Fig. 3. 18b) or side lying
FIGURE 3.18 (a) Stretching calf muscles in standing. E.nsure the knee is straight with hips extended.
(b) Stretching rectus femoris in prone. (c) Prolonged stretch to calf muscles on a tilt
table. Note: positioning the R leg on a stool ensures paretic limb is loaded.
(d) *Another way to stretch the calf muscles in standing that ensures the paretic limb
g; for is loaded. He also practises heel raise and lower in this position. (*Courtesy of K Schurr
gh to
and S Dorsch, Physiotherapy Department, Bankstown-Lidcombe Hospital, Sydney.)
Walking 107
y be
1 h
on In
~ ) .
rt of
s. At
(a) (b)
(c) (d)
108 Training guidelines
Hold stretch for approximately 20 s, relax, repeat 4-5 times.
Patients with severe muscle weakness who are physically inactive require
periods of prolonged passive stretching during the day.
Soleus: in sitting (see Fig. 4.14)
Gastrocnemius-soleus: in standing on a tilt table (Fig. 3. 18c)
Hold prolonged stretch for 30 min.
Ensure that the tilt table is as erect as possible. Backward leaning should
be avoided as it may interfere with regaining a perception of the vertical.
Eliciting muscle activity
Patients who are unable to produce sufficient muscle force to support the
body mass in standing may require intervention to elicit muscle activity in
lower limb extensor muscles. Simple exercises, such as the examples below,
may stimulate muscle activity in a simplified context. Equipment such as an
electrical stimulator, EMG biofeedback monitor and dynamometer may be
Electrical stimulation has a role in preserving muscle fibre contractility and,
since it can be done with minimal supervision, increases the time during
which muscles are being exercised. Augmented sensory feedback using an
EMG signal in either visual or auditory form may assist the patient to
activate weak muscles.
During isokinetic dynamometry, muscle activity may be elicited more easily in
the eccentric mode. Isokinetic eccentric training has been shown to be
superior to concentric training for the knee extensors (Engardt et al. 1995).
However, weightbearing through the limbs and practice of walking itself are
critical for promoting functional .muscle activity in the lower limbs.
Simple active exercise
Simple exercises may give the patient the idea of activating muscles of the
lower limb (e.g. hip extensors and quadriceps). Below are some examples
(Fig. 3.19).
Hip extension. Supine, affected leg over side of bed, practise small-range
hip extension by pushing down through heel (Fig. 3. 19a).
Knee at less than 90 flexion.
Discourage movement of the other leg and plantarflexion of affected leg.
If the foot slides forward, stabilize foot on the floor by pushing down
along the line of the shank.
Walking 109
FIGURE 3.19 Simple exercises to elicit muscle activity. (a) Therapist may need to stabilize foot to
prevent it from sliding forward. (b) Therapist flexes the knee to 90 and patient tries
to activate the hamstrings to lower the leg slowly (eccentric mode). It may be possible
to elicit an eccentric contraction before a concentric contraction.
(a) (b)
This exercise involves extensors of both hip (gluteus maximus and ham
strings) and knee (quadriceps). Hamstrings contraction, by its action at
the knee, also stabilizes the foot on the ground, preventing it from slid
ing forward. Encourage patient to sustain muscle activity by counting.
Incorporate muscle activity into loading the limb in standing.
This exercise is particularly useful for assisting the patient to achieve
alignment of the lower limb and pelvis in standing.
Patient sitting. Therapist extends the knee. Patient attempts to activate the
quadriceps to lower the leg slowly and repeat several times.
It may be possible to elicit an eccentric contraction before a concentric
Walking sideways
This exercise involves practice of a simple locomotor action in which the
number and extent of joint displacements to be controlled are reduced
compared to forward walking.
Starting with feet together, patient steps laterally while supporting and
n balancing the body mass on one leg then the other. SWing and stance: hips
should remain in neutral extension while stepping.
11 0 Training guidelines
Major angular displacements in the frontal plane are abduction and
adduction at the hips.
Foot placement should progress in a straight line laterally.
Placing toes along a line may give the patient the idea of this; however,
continually looking down at the feet should be discouraged.
A patient can practise walking sideways unsupervised or with minimal
supervision while resting hands on a raised bed rail (Fig. 3.20a).
Walking backward
Walking backward exercises the hip extensors particularly hamstrings.
FIGURE 3.20 (a) Walking sideways. Once she got the idea, she was able to practise semi-supervised.
(b) Walking backward.
(a) (b)
Walking 111
This exercise is only useful if the patient has some ability to flex the
knee while extending the hip (Fig. 3.20b).
The hamstrings act over two joints and must be critical to gait. For at least
two reasons little is understood about their actual function: (1) paucity of
investigation involving all three muscles in the group; (2) difficulty in inter
preting the electrical activity of two joint muscles in functional actions.
Strength training
There is evidence that muscle weakness is modifiable in individuals
following stroke (Bohannon and Andrews 1990, Sharp and Brouwer 1997,
Teixeira-Salmela et al. 1999) and in the elderly (e.g. Fiatarone et al. 1993,
Wolfson et al. 1993, Sherrington and Lord 1997). Increases in lower limb
strength are associated with improved walking speed, with the greatest
impact from strength increases in the weakest patients (Buchner et al. 1996).
There is increasing evidence that quadriceps muscle strength is critical for
dynamic stability in stance (Scarborough et al. 1999).
Emphasis in strength training is on lower limb muscles, in particular
extensors such as gluteus maximus, gluteus medius, hamstrings, quadriceps
and calf muscles, since these muscles provide the basic support, balance and
propulsive functions during stance. Abductors and adductors are major
muscles controlling lateral movement of the body mass. An increase in
strength, however, may not be sufficient to bring about improved walking
performance without practice of the action itself which is necessary for
neural adaptation to task and context to be regained (Rutherford 1988).
Amount and intensity of strength training (i.e. resistance, repetitions)
are graded to individual patient's ability.
Patient is encouraged to work as hard as possible.
As a general rule, a maximum number of repetitions should be per
formed, up to 10 without a rest and repeated three times. No increase in
spasticity with strength training has been reported (Sharp and Brouwer
1997, Teixeira-Salmela et al. 1999).
Functional weightbearing exercises
These exercises are designed to improve force generation, speed of muscle
contraction and synergistic muscle activity in the context of stance.
Simple loading of the affected limb
Generation of sufficient extensor force in the weightbearing limb is critical
to support of the body mass while the contralateral leg is raised and lifted
forward. The action in this exercise is similar to gait initiation (Brunt et al.
112 Training guidelines
FIGURE 3.21 Stepping with the non-paretic limb to load the paretic limb. (0) She has not moved
for enough forward (inadequate hip extension and ankle dorsiflexion). (b) A better
(a) (b)
Standing, stepping forward and backward with the contralateral limb
(Fig. 3.21). The body moss is moved diagonally and forward over the stance
limb to place the contralateral foot on a step.
Ensure that stance hip extends (approximately 15) and ankle dorsiflexes
(shank rotates forward over foot).
Step needs to be approximately 15 cm away from the toe of the stance
limb (depending on height of individual) to encourage hip extension and
forward movement of body mass. A footprint can be used to indicate
how far to step.
When stepping back, upper body should not flex forward over hip.
If body mass cannot be supported over foot without the leg collapsing
or the knee hyperextending, apply strapping or a splint to give some
support to the knee while exercising (see Fig. 1.6b).
Walking 113
In standing, movement of the body mass laterally to the stance limb is
preceded by movement of centre of pressure initially to the opposite
side before it moves to the stance side (Rogers and Pai 1990). Since this
exercise requires the coordination of forces produced by both legs, it is
practised to both sides.
Rapid stepping forward to place the non-paretic foot up on to a 15 cm
step placed 15 cm in front of the stance limb, without transferring all the
body weight, provides a particular challenge to mediolateral stability and
to gluteus medius (Sims and Brauer 2000). This is, therefore, an impor
tant exercise for increasing strength of the gluteus medius and hip stabil
ity, a problem following stroke and a known source of falls. Previous
studies have demonstrated that the hip adductors (Rogers and Pai 1993)
and superior portion of the gluteus maximus of the stepping limb can
also assist in control of the frontal plane movement.
Step-up/step-down exercises
These exercises train hip, knee and ankle extensors to work together, both
concentrically and eccentrically, to raise and lower the body mass. Step-up
exercises also train ankle dorsiflexors.
Forward step-ups (Fig.3.22a)
Foot ofaffeded leg on a step (approximately 8 cm high). Forward and upward
translation ofthe body mass over the foot with upper body remaining erect
Forces produced by the back limb move the body mass in a forward and
upward direction over the forward foot (ankle dorsiflexion). Extension
of hip, knee and ankle of the affected limb raises the body mass as the
back foot is lifted on to the step.
When stepping back down, flexion of hip, knee and ankle of support leg
lowers the body mass until the foot touches the ground.
Therapist places the foot on to the step if necessary.
Ensure that the shank rotates forward at the ankle (dorsiflexion). This
brings the body mass forward over the foot.
Shank and foot may need to be stabilized to provide a stable shank over
which the knee can extend and prevents knee hyperextension.
Major extensor force generation of support limb is relatively evenly dis
tributed over the three lower limb joints in forward step-ups (Agahari
et al. 1996).
~ s
Lateral step-ups (Fig. 3.22b)
Foot of affected leg on a step. Step up with unaffected limb with body mass
remaining erect.
g Extension of hip, knee and ankle of affected limb raises the body mass,
e and adduction of hip moves the body mass laterally as the contralateral
foot is lifted on to the step.
114 Training guidelines
FIGURE 3.22 Step-ups: (a) forward; (b) lateral; (c) step-downs. (d) The therapist holds her foot to
stop external rotation as she steps down. Note (a) and (b) can be practised in a
(a) (b)
(c) (d)
Walking 115
When stepping down, flexion at hip, knee and ankle lowers the body
mass until the foot touches the ground.
Therapist places the foot on the step if necessary.
Major extensor force generation of the support limb in lateral step-ups
is concentrated at the knee with some involvement at the ankle, and
strong adductor/abductor forces at hip (Agahari et al. 1996).
Forward and lateral step-ups can be practised in a harness.
Forward step-downs (Fig. 3.22c)
Both feet on step, step down with the unaffected leg.
Body mass remains supported and balanced on the affected limb which
flexes at hip, knee and ankles to lower body mass for foot placement on
to the floor.
When stepping up backward, hip, knee and ankle of the support limb
extend to raise body mass while foot is lifted back on to step.
Force is also produced by the front limb to move body mass backward
and add propulsive force to the upward movement.
Do not allow the support (paretic) limb to externally rotate during step-
downs (Fig. 3.22d). This happens if soleus is stiff or contracted.
Progress step-up and step-down exercises by increasing the height of the
step, number of repetitions and speed, by stepping without stopping (i.e. no
foot placement) (Fig. 3.23), and by practising reciprocal stair walking
(average riser height 17 cm).
Heels raise and lower
Since the major power generator during the gait cycle is ankle
plantarflexion during push-off (Winter 1987) every effort is made to
strengthen and train plantarflexor activity. This pre-swing muscle activity
accelerates the lower limb into swing phase. During stance phase, ankle
plantarflexors also contribute to ankle and knee stability and to the
control of forward translation of the body mass over the foot.
Forefeet on step, heels free (Fig. 3.24). Heels are lowered as for as possible
then raised to plantigrade.
Hips and knees remain extended throughout. Knee flexion may indicate
that patient is having difficulty coordinating synergistic activity of knee
extensors which contract to counter flexion of the knee by two-joint
Ensure weight is over affected leg.
Running shoes should be worn.
116 Training gUidelines
FIGURE 3.23 Repetitive semi-supervised continuous lateral step-ups. Note the paretic limb is on the
edge of the block to prevent foot placement. (Courtesy of K Schurr and S Dorsch,
Physiotherapy Department, Bankstown-Lidcombe Hospital, Sydney.)
Increased stiffness in ankle plantarflexors may be largely due to mechan
ical changes in muscle fibres (e.g. Dietz et al. 1981).
As strength increases the exercise is done on one leg (Fig. 3.24b).
Increased stiffness and contracture of calf muscles commonly have a
negative effect on function. Maintaining length and flexibility of these
muscles is a priority.
This exercise places an active stretch on the plantarflexors as the heels
are lowered (eccentric contraction). It also gives the idea of activating
the plantarflexors (concentric contraction) from a lengthened position
similar to that of push-off at end of stance.
It is clinically evident that this exercise decreases stiffness in the
calf muscles. There is no evidence that the exercise increases
Walking up a ramp is another exercise for stretching and strengthening
calf muscles (Fig. 3.25).
Walking 117
FIGURE 3.24 (a) Heels raise to plantigrade and lower to strengthen and stretch plantarflexors. This
exercise trains the calf muscles specifically at the length required for push-off at end
of stance. Patient is holding on to a stable object. (b) With improvement in strength
and skill, the exercise is performed on one leg.
(a) (b)
Non-weightbearing strength training
These progressive resistance exercises are designed to improve force
generation, speed and sustainability of muscle contraction (Fig. 3.26).
However, as Scarborough and colleagues (1999) comment, integrating
strength changes into meaningful function needs practice of the action itself.
Isokinetic dynamometer
Elastic band resistance exercises
These exercises can be done as part of circuit training.
Encourage patient to do a maximum number of repetitions, up to 10,
without stopping. Repeat three times. Aim to train at 80% maximum
isometric voluntary contraction (Ch. 7).
Muscles apart from the target group are also active as synergists (e.g. as
stabilizers). Some increase in strength in these muscles also can be
expected (Fiatarone et al. 1990).
Increase resistance on the dynamometer by upgrading to the next rub
ber band. Bands are coloured, each colour representing a certain ten
sion, yellow has the least tension, black the most (Ch. 7).
118 Training guidelines
FIGURE 3.25 Walking up a ramp. She needs to be encouraged to use her calf muscles to push off.
FIGURE 3.26 Some examples of semi-supervised exercising. (a) The MOTOmed provides resistance
or assistance in response to the patient's performance (Reck, Reckstrasse 1-1,
0-88422 Betzenweiller, Germany). (b) Exercising on the Orthotron to strengthen knee
extensor muscles concentrically and eccentrically or isometrically (Cybex Human
Performance Rehabilitation, 2100 Smithtown Ave, Ronkon Koma, NY 11779, USA).
(a) (b)
Walking 119
FIGURE 3.26 (c) The elastic band should be under tension at all parts of the range.
Maximizing skill
Walking involves a basic locomotor pattern which is normally adapted
in the presence of obstacles such as kerbs and steps, speed demands
and other environmental factors. The opportunity for challenging and
variable practice is critical to optimize performance in everyday terrains
and contexts. Variable practice (Fig. 3.27) involves walking in a busy
corridor, walking at speeds required to cross a street safely, practice of
stepping on and off moving walkways and escalators, walking through
automatic doors, under and over obstacles, up and down ramps and
kerbs, under different lighting conditions. The patient needs practice in
identifying potential threats to balance and making the appropriate
avoidance strategies. The ability to increase walking speed to avoid
a moving obstacle is an important safety issue. A recent study has
demonstrated the difficulty patients have in stepping over obstacles
of different heights and widths (Said et al. 1999). Patients need to
practise coping with this common everyday hazard (Fig. 3.28). Turning
while walking, stopping, and reciprocal stair walking all need to be
Increasing aerobic capacity has been found not only to improve
cardiorespiratory function in stroke patients but also to increase functional
ability (Ch. 7). Treadmill walking, bicycle ergometry and adapted cycle
ergometry are used to increase fitness.
Walking 121
Walking over obstacles can be started once the person can walk independently. This
figure illustrates some common difficulties. (a) She steadies herself with the rail, and
has positioned her L foot so only a small step is needed. Note the path of the R limb
R hip abducts excessively, hip flexion and ankle dorsif'exion are reduced. (b) She has
increased step length and can clear the obstacle without needing support. (c) She
catches her heel on the f'exib'e obstacle. Poor placement of supporting (trail) foot
may contribute to this. Practice should include walking over wider obstacles requiring
both limbs to step over. Obstacle crossing provides practice of balancing on one leg,
controlling the body mass as it moves forward and controlling step length and foot.
(a) (b) (c)
An important concern for a therapist is to determine which aspects of perfor
mance should be measured to make a valid assessment of progress.
Observational analysis gives uncertain results. Valid and reliable tests enable
the documentation of the patient's progress and the effects of intervention.
Standardized protocols should be followed in administering all walking tests.
Even the level of encouragement provided by a test supervisor to subjects
undergoing walking tests can have a large effect on their performance (Guyatt
et al. 1984). The tests listed below have been shown to be reliable if performed
under standardized conditions.
Walking velocity may be the most important objective clinical measure of
functional ability. It is simple to administer and is easily understood and
accepted by patients. Almost all other gait measures are speed dependent
(Andriacci et a1. 1977); for example, as speed increases so does the extent of
hip extension and amount of push-off. Another critical measure involves the
testing of walking ability outside the house, within the community.
Functional tests
These include:
10m walk test
6- or 12-minute walk test (McGavin et al. 1976, Guyatt et al. 1985)
Motor Assessment Scale: Walking item (Carr et al. 1985)
Timed 'Up and Go' Test (TUG) (Podsialo and Richardson 1991)
Step Test (Hill et al. 1996)
Obstacle Course Test (Means et al. 1996)
10m walk. The subject is timed while walking between two marks on the
floor, using a stopwatch. Velocity is calculated, as distance/time, in mls. To
avoid the effects of acceleration and deceleration, the subject is timed only over
the middle 10m of a 14 m walkway. The 10m walk test is useful in the early
stages as a guide to progress. It should not be taken to reflect functional walk
ing as it involves a very short distance. This has been confirmed recently (Dean
et al. 2001).
Other clinically feasible spatiotemporal measures that yield valid quantltlve
data include include step and stride length, stride width and cadence. Stride
and step lengths are measured using foot switches, an instrumented walkway
or a Stride Analyser". Attaching felt pens to the heels can provide a simple and
inexpensive method of collecting these data.
The 6- or 12-minute walk test measures the distance walked in 6 or 12 min
utes. These tests are used extensively in individuals with cardiorespiratory
conditions. The 6-minute test is particularly useful since there are some norma
tive data available for comparison (Enright and Sherrill 1998). This test may
provide a more realistic appraisal of function and enable an accelerated pro
gramme of training to prepare the patient for independent life after discharge.
TUG. Patient stands up from a chair, walks 3 m, turns around, returns to
chair and sits down. Time taken to complete the test is measured with a
stopwatch. Able-bodied subjects are able to perform this test in less
than 30s.
Step test is used to evaluate the ability to support and balance the body mass
on the affected lower limb while stepping with the unaffected limb. Starting
with feet parallel and 5 cm in front of a 7.5 cm high block, patient steps for
ward and backward repeatedly as fast as possible for 15 s. The number of steps
is counted.
Where a biomechanical laboratory is attached to the rehabilitation centre,
comparative biomechanical tests can be used to evaluate progress.
*B & L Engineering, 3002 Dow Ave, Suite 416, Tustin CA 92780, USA.
Biomechanical tests
Variables to be tested include:
angular displacements
ground reaction forces
moments of force, power, energy.
Matching observations with biomechanical test results enables physiotherapists
to get feedback on and improve their observational skills.
Calf muscle length test
A reliable method of measuring passive ankle dorsiflexion in a clinical setting
is described by Moseley and Adams (1991). It involves the use of skin surface
markers (head of 5th metatarsal, lateral malleolus and head of fibula), polaroid
photography and the application of a known torque, using a specially designed
template. The knee is held at a constant angle in extension with a velcro strap
over the shank (see Fig. 7.4).
Physiological tests
Physiological tests have shown that speed and energy costs during gait are
valid parameters in assessment of walking capacity. Fitness is typically tested
using a treadmill, bicycle ergometer or MOTOmed Pico Leg Trainer. Fitness
tests are described in Chapter 7. The fitness test most specific to gait uses the
treadmill. Monitoring heart rate is a simple method of ensuring that exercise is
sufficiently vigorous.
Treadmill training
The rationale for considering treadmill training in rehabilitation of gait follow
ing stroke arose from animal research investigating motor control mechanisms
in lesioned animals (Grillner and Shik 1973). There have been several recent
clinical studies showing the positive effects of treadmill training on individuals
with stroke, although such training has yet to be widely used in clinical set
tings. Several studies have demonstrated effects superior to those obtained by
neurodevelopmental therapy (Bobath) intervention (Fig. 3.29) (Richards et al.
1993, Hesse et al. 1995, Pohl et al. 2002). There is no evidence that treadmill
training causes 'mass synergies' or increases spasticity (Hesse et al. 1995, Hesse
1999), a concern which may be expressed by clinicians.
Many of the gait deviations seen in stroke patients result from their inability to
adequately bear weight through the affected lower limb during the loading
phase of the gait cycle (Carr and Shepherd 1998). Treadmill training using a
FIGURE 3.29 Means and standard deviations of walking velocity over time. Walking velocity
increased during treadmill training (A 1, A2) but did not increase during conventional
therapy. B, Bobath. (Reproduced from Hesse et al. 1995, with permission.)
A1 B A2
~ 0.4
0.0 I , I I , I I, I I I I I I, I I I , I
1 2 3 4 5 6 7 8 9 10 11 12131415161718
harness for partial body weight support (PWS) seems to address this problem
and can allow a patient with severe weakness of lower limb extensor muscles
to go through the gait cycle with some assistance from the therapist. The
harness also provides a sense of security for the patient who need not fear
Hesse and colleagues (1997) investigated the influence of various amounts of
PWS compared with full weightbearing. Their data suggest that 30% PWS
results in the most normal gait. A recent randomized clinical trial involving
stroke subjects reported better outcomes for the experimental group who
trained with PWS (decreased as subjects' gait improved) compared with sub
jects who trained on the treadmill bearing full weight on the lower limbs
(Visintin et al. 1998). At the end of 6 weeks, 79% of PWS subjects were train
ing with full weight. This group scored significantly higher than the non-PWS
group on overground walking speed and endurance and on balance using the
Berg Scale (Table 3.1).
The major benefits from training on a treadmill with PWS include the following.
Task-specific gait training can be started early after a stroke even before
the patient is able to support 100% of the body mass on the affected lower
The patient can practise the complete gait cycle, i.e. support, balance and
Inter- and intralimb timing parameters are practised before the patient has
sufficient muscle strength to fully support body weight.
Loading the affected limb can be varied according to ability to support
The harness eliminates the need to use adaptive movements that favour an
asymmetrical gait.
Speed of the treadmill can be increased, thereby stimulating the patient to
walk faster.
If performed intensively (progressively increasing speed, slope and duration),
without PWS, treadmill walking can increase muscular endurance and
cardiovascular fitness (Ch. 7).
Treadmill walking represents a biomechanical stimulus that may lead to
improved motor control. There are many similarities between treadmill and
overground walking (Strathy et al. 1983, Murray et al. 1985, Arsenault et al.
1986). A biomechanical study investigating the equivalence of treadmill walk
ing with PWS and overground walking in healthy subjects at different speeds
and different percentages of support reported only minor differences (Finch
et al. 1991). The major differences between treadmill walking and overground
walking are in the visual and kinesthetic information perceived by the walker.
Therefore, once a patient is able to walk unaided, training incorporates walk
ing in more natural environments and for different goals.
Certain characteristics of treadmill walking are probably beneficial to the
training of gait by externally forcing a rhythmical and dynamic gait pattern.
The moving belt 'forces' hip extension and ankle plantarflexion in support
phase. With the treadmill moving at a reasonable speed, the relatively strong
)lem stretch applied to the hip flexor and calf muscles at the end of stance may
cles enhance muscle activity for swing.
fear As Hesse (1999) points out, further comparative studies are needed to meet the
criteria for evidence-based practice (e.g. large randomized controlled multicen
tre trials). Meanwhile treadmill training in neurorehabilitation has a sound
:s of scientific rationale, and many clinical studies have compared its efficacy with the
lack of efficacy of conventional intervention in the training of gait. We suggest
vmg that treadmill walking is likely to be maximally effective when combined with
who functional lower limb strength training and practice of walking overground.
Electrical stimulation
Electrical stimulation (ES) has had a long and varied history in rehabilitation
. the
post-stroke. There is evidence that ES has the potential for positive effects on
muscle. These include stimulating muscle fibre contractility, increasing strength
(Glanz et al. 1996), decreasing contracture (Hazlewood et al. 1994) and pre
venting a build up of connective tissue (Williams et al. 1988).
wer Early in stroke, while the patient is relatively inactive, ES can be used as an
adjunct to exercise to maintain the integrity of muscles. It should be given in
and particular to those muscles likely to adapt negatively to physical inactivity (e.g.
calf muscles, quadriceps, tibialis anterior) as a means of stimulating muscle
fibre contractility. ES units are relatively inexpensive, extremely durable and
reliable and can be used by the patient or a family member after instruction.
A meta-analysis (Glanz et al. 1996) of four randomized controlled trials testing
the effect of functional electrical stimulation (FES) on muscle force post-stroke,
supports its efficacy in improving muscle strength. Studies of FES-assisted gait,
involving the use of multichannel stimulators (e.g. Malezic et al. 1987, Bogataj
et al. 1989), have also reported to improved aspects of gait performance.
Whether or not either of these effects is lasting or transfers into clinically sig
nificant improvement is so far unclear.
Aids and orthoses
An orthosis is defined by the International Society for Prosthetics and Orthotics
as an externally applied device used to modify structural and functional char
acteristics of the neuromuscular system. The most commonly prescribed ortho
sis with a view to improving gait is an ankle-foot orthosis (AFO).
A major consideration in determining the use of an orthotic device is that the
device, by constraining motion at one joint, alters body mechanics which necessi
tates adaptations at other joints. These adaptations may increase energy cost
during gait. One study compared a group of able-bodied subjects who wore either
a narrow or wide semi-rigid plastic AFO set in neutral position (Balmaseda et al.
1988). When walking with either AFO there was a decrease in the duration of
stance phase of the ipsilateral leg, centre of pressure was shifted to a more lateral
position throughout stance, and the point of impact at heel contact was more
posterior. In these subjects the magnitude of vertical force at toe-off was
increased, indicating that more muscle force was required to overcome the resis
tance of the AFO and propel the body mass forward. An AFO has, therefore, the
potential to impact negatively on aspects of gait performance.
Research into the potential of AFOs to control foot/ankle mechanics during
gait, and thereby improve gait performance following stroke, has equivocal
results and the issue of whether or not to prescribe some form of AFO remains
controversial. There are several descriptive studies comparing the effects of dif
ferent types of AFO on aspects of gait after stroke (Lehmann et al. 1987,
Burdett et al. 1988, Beckerman et al. 1996). However, for methodological and
theoretical reasons it is difficult to draw specific conclusions from these studies.
One of the problems in evaluating this literature is the theoretical assumption
that it is calf muscle spasticity that is impeding gait, and that the AFO will
affect foot position. Where the term spasticity is used in a generic sense, it is
not clear whether the problem is soft tissue contracture, increased muscle stiff
ness or hyperreflexia.
Rigid or semi-rigid AFOs are often prescribed for patients with 'foot drop'.
Both anterior tibial muscle weakness and contracture of calf muscles can inter
fere with ankle dorsiflexion during swing phase, impeding toe clearance and
heel contact. Although a rigid AFO holds the foot in some dorsiflexion during
swing, it imposes a mechanical resistance to plantarflexion for push-off. Use of
. ~ e ,
a muscle stimulator to produce anterior tibial muscle activation, strengthening
exercises and stretching for calf muscles are likely to be more effective for such
a problem and enable a more energy-efficient gait.
From the evidence so far, hinged AFOs that allow dorsiflexion and/or plan
tarflexion are preferable to fixed orthoses. Ankle orthoses that provide medio
lateral stability at the subtalar joint without resistance to dorsiflexion or
plantarflexion, such as an Air Stirrup Brace", may be useful for patients whose
gait is impeded by such instability (Burdett et al. 1988).
Until there is more research that addresses these issues, we suggest that the
essential prerequisites in prescribing an AFO, should it be considered neces
sary, are that:
biomechanical gait mechanisms underlying gait dysfunction have been
relative contribution of active and passive soft tissue components has been
orthotic device fits, neither allowing unwanted movement nor restricting
desired motion, and can be shown to improve the effectiveness and
efficiency of gait for that individual.
Where plantarflexor muscles are very stiff, hyperreflexic and/or there is soft tis
sue contracture, low-dose botulinum toxin or serial casting combined with walk
ing training may be necessary. An AFO is unlikely to produce a measurably more
effective gait under these conditions. Reiter and colleagues (1998), using a com
bination of low-dose botulinum toxin and ankle-foot taping, reported a decrease
in Ashworth scores and increases in gait velocity and step length.
Walking aids
Walking aids - parallel bars, walking frame, threelfour point stick (tripod,
quadrupod), stick, an assistant's arm - have a number of predictable draw
backs. They unload the weightbearing structures of the lower limb and can
increase muscle activity and support in the upper limb (Winter et al. 1993).
They may compromise an already reduced gait velocity.
Walking sticks or canes have been prescribed following stroke for many years
although their efficacy in increasing stability and improving gait has not been
documented. It is only recently that the effects of different types of sticks have
been compared. A quadrupod stick offers no advantage over a standard stick
in terms of standing balance (Milczarek et al. 1993) or in spatiotemporal vari
ables in walking (Joyce and Kirby 1991). In a comparative study of stroke
patients walking with and without a stick there was no increase in speed with
the stick compared with no stick (Kuan et al. 1999). However, step and stride
length with the affected leg were increased and cadence and stride width
decreased when the stick was used, suggesting that a stick may have a greater
effect on spatial variables (stride length and stride width) than on temporal vari
ables (speed and gait phases). A recent study found that individuals post-stroke
*Aircast Inc., PO Box T, Summit, NJ 07901, USA.
walking with stick assistance relied mostly on the unaffected limb for propul
sion, while using the affected limb and stick for braking (Chen et al. 2001).
Walking with a pick up walking frame is constrained by the structure of the
frame. There are periods of motion and of stasis with the result that gait veloc
ity is markedly decreased and hips remain flexed (Crosbie 1993). Roller walk
ers, if they are high enough to prevent hip flexion, may be a better option for
the very frail individual, who should be encouraged to take longer steps and to
increase the distance walked each day.
Parallel bars are designed for arm support with partial weightbearing through
the lower limb(s). They are, therefore, useful for training patients with spinal
cord injuries who have to rely on upper limbs for support and balance.
Although sometimes used in stroke rehabilitation, walking in parallel bars has
major disadvantages, encouraging use of non-paretic limbs and preventing the
regaining of appropriate balance mechanisms.
Propelling a one-arm drive wheelchair with the intact arm and leg may promote
early independence in getting from one place to another. However, it focuses
attention on the intact side and is likely to contribute to learned non-use.
In summary, although all walking aids impose some mechanical constraint, a
simple walking stick interferes least with balance and walking yet provides
some stability (Kuan et al. 1999). Further research on the effects on gait of dif
ferent types of orthoses and walking aids would assist clinicians to prescribe an
aid, or not, with some understanding of the potential mechanical and physio
logical effects.
Strength training
Maximizing skill
Functional tests
Biomechanical tests
Augmented feedba(jk
Analysis of motor performance
Research findings
Observational analysis
Guidelines for training
Standing up sitting down
Inability to stand up is common early on following stroke and difficulty per
forming the action may continue beyond discharge, limiting independence and
participation in everyday life. Inability to stand up independently predisposes the
individual to further decreases in muscle strength and physical fitness, and to
adaptive soft tissue changes, particularly in soleus muscle, associated with disuse
and physical inactivity.
Difficulty standing up is reported to be a common source of falls (e.g. Sorock
and Pomerantz 1980, Yoshida et al. 1983, Tinetti et al. 1988), with a report
that 20% of falls occurred while getting out of a wheelchair and 22 % while
getting out of bed (Sorock and Pomerantz 1980). The tendency to fall during
attempts at standing up is not surprising given the deficits in muscle strength
and postural stability following stroke, and lack of independence in standing
Standing up and
sitting down
The ability to stand up (STS) and sit down (SIT) is essential to an independent
lifestyle. Standing up is a prerequisite for other actions such as overground and
stair walking which require the ability to get into the standing position. Its inci
dence in daily life is high, an early study reporting that able-bodied individuals
stand up on average four times an hour (McLeod et al. 1975).
Standing up is one of the most mechanically demanding daily activities, requir
ing greater range of motion at the knee and higher moments of force at both hip
and knee than gait or stair climbing (Andriacchi et al. 1980, Berger et al. 1988).
Lower limb muscle strength, balance and control of the total body segmental
linkage are therefore critical for effective performance.
up has been found to be one of the most likely factors associated with an
increased risk of institutionalization (Branch and Meyers 1987).
Despite the significance of this action to daily life, until recently there has been
little focus in clinical practice on the specific training of STS and even less on SIT.
We have argued for some time (Carr and Shepherd 1987, 1998) the importance
of intensive practice designed to improve the performance of STS/SIT, together
with methods of increasing lower limb muscle strength and weightbearing
through the paretic limb. Over the last decade there has been more clinical focus
due in large part to an increasing number of biomechanical studies reporting the
principal dynamic features of the action and some clinical studies showing that
training can be effective in improving motor performance (see Table 4.1).
Standing up
The majority of biomechanical studies have focused on movement in the sagittal
plane, using two-dimensional video analysis systems with subjects standing up
with feet on one or two force plates. Results of investigations performed in the
laboratory provide biomechanical data which are used to guide the development
and testing of training protocols. These data also enable a rational analysis of a
patient's performance. Without such information about the dynamic characteris
tics of the action and the effects of different task and environmental conditions,
the therapist can only guess at what constitutes 'normal' movement.
STS requires translation of the body mass in a horizontal and vertical direction
from a relatively stable sitting position with thighs and feet as base of support, to
a period of relative instability when the thighs leave the seat and the feet become
the base of support (Fig. 4.1). Generating the angular and linear momentum to
perform the horizontal to vertical translatory movements of the body mass is
potentially destabilizing. Postural stability is most threatened around the time the
thighs lift off the seat. At this point, angular displacement of the relatively large
upper body (head, arms and trunk) pivoting at the hips changes from flexion to
extension, and linear momentum of the body mass changes from horizontal to
vertical. Braking forces are required to control horizontal momentum, ensuring a
stable transition from horizontal to vertical movement. Translation of the body
mass forward depends on strength and control of muscles crossing hip, knee and
ankle. These muscles generate the necessary forces to support and balance the
body mass while at the same time propelling it vertically. In everyday life, such
factors as foot position and weight distribution between the two feet vary, adapt
ing to task and environmental demands.
For ease of description, STS can be divided into a pre-extension and an exten
sion phase (Carr and Gentile 1994, Shepherd and Gentile 1994), the transition
occurring at thighs-off (TO). Although other methods of identifying phases of
the action have been reported, this may be the simplest division for clinical pur
poses. In reality, the pre-extension and extension phases form one continuous
Standing up and sitting down 131
FIGURE 4.1 Standing up. Horizontal movement of the body mass: forward rotation of the trunk
at the hips and of the shonk at the ankles. Vertical movement: extension of hips,
knees and ankles.
translatory movement of the body mass horizontally, changing to movement
vertically without a pause (Fig. 4.2).
Kinematics and kinetics
In the pre-extension phase the feet are placed well back so that when the upper
body is rotated forward at the hips, lower limb muscle activity and segmental
movement generate the posteriorly directed component of ground reaction
forces, thus propelling the body mass forward. Biomechanical investigations
of STS typically model the pelvis and spine as one segment called 'trunk' in order
to capture the action of the upper body as it rotates forward. These segments
act together functionally as a 'virtual' segment with minimal motion occurring
within the spine itself and between the spine and pelvis. Rotation forward of this
virtual segment by flexion at the hips, together with rotation of the shank at the
ankle by dorsiflexion, move the body mass forward over the feet. That is, the
trunk and shank segments act as a functional unit for bringing the body mass
forward (Pai and Rogers 1991). In the extension phase, vertical movement is
brought about by extension at hips, knees and ankles in a typical sequence of
onsets, the knees starting to extend (at TO) before the hips and ankles (Shepherd
and Gentile 1994).
Since the feet are on the supporting surface, horizontal and vertical ground
reaction forces (GRFs) play an important part in the movement. Prior to TO,
horizontal GRFs generated in a posterior direction (approximately 10% body
weight) serve to propel the body mass forward. These are followed by anteri
orly directed forces to brake the forward momentum. At TO, there is a rapid
rise in amplitude of vertical GRFs to a peak of approximately 150% of body
weight depending on factors such as speed (Fig. 4.3). These forces decrease
132 Training guidelines
FIGURE 4.2 Superimposed stick figures graphed from x/y coordinates. Shoulder and knee paths
can be seen. Note knee moves forward early in movement (ankle dorsif'exion and
knee f'exion) and then backward as hips, knees and ankles extend. Final standing
alignment: hip marker is in front of ankle marker (lateral malleolus).
FIGURE 4.3 Typical vertical (z) ground reaction force profile of an able-bodied subject standing up
at a preferred speed of approximately 1.5 s. Vertical line indicates thighs-off. Arrows
indicate movement start and end.
x = 1.18
.5000 1.500 2.500 3.500 4.500
Time (5)
after TO and stabilize at body weight when standing is reached. This pattern
reflects the requirement to generate force greater than body weight to acceler
ate the body vertically.
The major extensor muscle forces occur therefore at TO. As in the stance phase
of gait, peak extensor moments of force at individual joints can vary coopera
tively. However, for each limb overall, support moment (an algebraic summa
tion of hip, knee and ankle moments of force) of approximately four times the
body mass peaking at TO, propels the body mass vertically (Shepherd and
Gentile 1994).
STS under different conditions
Identifying the fundamental biomechanical determinants of STS enables the clin
ician to organize training with a greater understanding of the major dynamic
interactions occurring within the segmental linkage. Certain biomechanical vari
ables under constrained conditions across different age groups have been found
to be remarkably consistent and repeatable within and between subjects, for
example, angular displacements, sequence of joint onsets, linear paths of body
parts and support moments of force.
However, for daily life, flexibility of performance is essential. Normally we
adapt the action to the prevailing internal (pregnancy, joint pain) and external
(seat type, task) constraints. For example, the extent of angular rotation varies
depending on height of stool in relation to leg length, extent of thigh support
and type of chair. Investigations of several external conditions of relevance to
clinical practice (foot placement, initial trunk position, speed of movement,
seat height) have been found to affect the dynamics of the action in ways that
could impact negatively for a patient with motor system impairments including
muscle weakness. Understanding these effects makes it possible to set up opti
mal practice conditions for patients who are having difficulty performing the
action effectively.
Foot placement. Several studies have investigated the mechanical effects of dif
ferent foot placements (Fleckenstein et al. 1988, Shepherd and Koh 1996), i.e.
the angles at ankle and knee prior to the start of the action. Prior to standing
the feet are normally drawn back so that they are placed posterior to the
knees. The position of the feet affects the distance the body mass has to be
moved forward. Laboratory studies have shown that the further forward the
feet, the greater the magnitude of hip flexion and the greater the velocity
required to overcome the potential braking force created by the anterior foot
position (Vander Linden et al. 1994) and to move the body mass the additional
distance forward (Shepherd and Koh 1996). Anterior foot placement was
shown to be associated with an increase in peak hip moments of force and an
increase in the duration of the extension phase, with muscle force therefore
being sustained for a longer period than when the feet were back (Vander
Linden et al. 1994, Shepherd and Koh 1996).
The clinical implication from these studies is that for ease of standing up the
most biomechanically effective foot position is with the ankles at approximately
Angular momentum produced by the trunk as it rotates forward at the hips is
the major contributor to horizontal momentum of the body mass (Pai and
Rogers 1991) and appears to facilitate the lower limb extension that raises the
body mass to the standing position. A study in which the initial trunk angle at
Timing and speed oftrunk rotation. Several studies have described the contribution
of trunk flexion to STS (Pai and Rogers 1990, Schenkman et al. 1990, Shepherd
and Gentile 1994). It is evident that a timing relationship between flexion of
the trunk at the hips and onset of lower limb extension may be a critical fea
ture in the movement's organization. The extension phase of STS starts at the
knee while the trunk is still flexing at the hip. Peak hip angular acceleration
has been found to occur simultaneously with the onset of knee extension. This
means that the body mass starts moving upward while it is still moving for
ward. There is a smooth continuous transition from pre-extension to extension
phase that occurs as a result of intersegmental coordination. This mechanism
of momentum transfer is probably a major factor in conserving energy. It is
possible to stand up from a position with the upper body already forward at
the hips, but as one of the following studies shows there is a cost in terms of
additional extensor muscle force.
75 dorsiflexion (Fig. 4.4). For this to be possible after stroke it is necessary to
preserve the extensibility of soleus muscles by both passive and active stretch
ing. Repetitive practice of STS with the feet placed back is itself a method of
providing active stretch to soleus.
Preferred foot placement is, on average, 10 cm from a vertical line drawn from
the middle of the knee joint (approximately 75 dorsiflexion). Distance was
calculated from Shepherd and Koh 1996.
the start of STS was varied from erect (0) to flexed (60) provides evidence in
support of this mechanism. The results indicated that a high level (more than
three times body mass) of support moment of force over the three lower limb
joints was produced for longer when subjects stood up from the 60 position,
i.e. without the facilitatory effects of active hip flexion, than when the move
ment started with the trunk erect (Shepherd and Gentile 1994).
Speed of movement also impacts on the extension phase. In a study in which
subjects stood up at three different speeds (slow, preferred, fast), it was evident
that increased hip flexion velocity when subjects moved fast had a potentiating
effect on the production of lower limb extensor force. When subjects moved
slowly, however, a significantly longer period of time was spent producing a high
level of support force, illustrating the effect of reduced velocity of trunk move
ment, and therefore momentum, on the production of lower limb muscle forces
(Carr et al. 2002).
The findings from these studies show that if patients do not rotate the upper
body segment through a range of 30-40, at a reasonable speed, and do not
move the body mass forward and upward without a pause between the two, they
will have to work harder to stand up. The clinical implications are that active
rotation of the upper body should start with the upper body erect, and there
should be no pause between the pre-extension and extension phase. These condi
tions may optimize the transfer of momentum from horizontal to vertical and
potentiate lower limb extension.
Seat height"'. Studies designed to investigate the effect of different seat heights
on moments of force at lower limb joints have found that seat height can have
significant effects on performance (Burdett et al. 1985, Rodosky et al. 1989).
Rodosky and colleagues reported a greater than 50% decrease in the moment of
force at the knees from the lowest to the highest seat height. There was a rela
tively small change in peak moments at the hips, while peak moments at ankles
were unaffected by chair height (Fig. 4.5). As chair height was increased peak
angular displacements at hip, knee and ankle were also decreased.
The clinical implications from this work are related to training methods and
selection of chair type for individuals with difficulty getting into standing.
Raising seat height enables an individual with weak muscles to practise STS and
SIT. As strength increases, seat height is lowered, providing progressive strength
training for lower limb extensors.
Important factors in chair type to assist STS are, therefore, the height of the
seat and the absence of any structural block to posterior foot placement at
*During STS the hips are flexed when the largest hip moments are being produced (Andriacci et al.
1980). In individuals with joint disease or endoprostheses, the stresses created may accelerate
hip damage or loosen a prosthesis (Fleckenstein et al. 1988). STS from a relatively high chair
with feet placed back decreases the magnitude of hip flexion required to get the body mass
forward. Such individuals may also need to use arm rests for assisting STS in order to decrease
the hip moments.
FIGURE 4.5 An illustration of the change in peak moments of force occurring with different
choir heights at hips, knees and ankles. (Reproduced from Rodosky et 01. 1989, with
12.0 I
~ !
I 8.0 i
>< ~
~ i
'Cf 6.0 -j

H- Hip
K - Knee
2.0 i
A- Ankle
0.0 i
80% 100%
hi ~
Chair height
(% knee height)
approximately 75 dorsiflexion. Seats need to be high enough to enable the indi
vidual to stand up without gross adaptation (e.g. most of the weight through
the non-paretic limb). A survey in which elderly people were asked to rank the
five factors they considered important in terms of chair design found that 'Easy
to get out of' was the most important factor when choosing a chair. 'High seat'
was ranked third, before comfort (Munton et al. 1981). It is notable that a chair
specially designed for comfort in sitting with a deep, posteriorly slanted seat
was found to involve greater quadriceps activity and greater hip and knee flex
ion, and therefore appeared more difficult to stand up from, than a standard
armchair (Wheeler et al. 1985). Recommendations for chair type must therefore
take into account both ease of standing up and comfort (Fig. 4.6).
Muscle activity
The principal muscles in the pre-extension phase are hip flexors and ankle dor
siflexors to move the body mass forward. There is some isometric contraction
of spinal and abdominal muscles to stabilize the extended trunk. Rectus
abdominus activity has been found to be minimal and sporadic (Millington
et al. 1992, Vander Linden et al. 1994). Hip, knee and ankle extensor muscles
propel the body mass vertically in the extension phase.
Electromyographic (EMG) studies (Fig. 4.7) have shown that tibialis anterior is
activated early in the action, reflecting its contribution to foot placement
~ <easy to get out or chair with adjustable seat height.
Illustration of mean muscle onsets and duration of activity from able-bodied subjects
standing up at their preferred speed. Left and right error bars indicate standard error
~ a t '
of onset and duration time respectively. TA, tibialis anterior; RF, rectus femoris; 8F,
biceps femoris; VL. vastus lateralis; GAST, gastrocnemius; SOL, soleus; 0"", movement
onset; 31"", thighs-off; 100%, movement end. (Reproduced from Khemlani et 01. 1998,
with permission.)
40 50 60 70 80 90 100 110 120
. is Percentage of total movement
-20 -10 o 10 20 30
~ n t
backward, stabilizing the shank and foot, and moving the body mass forward
(Khemlani et al. 1998). Onsets of hip and knee extensor muscles (gluteus max
imus, biceps femoris, rectus femoris, vastus lateralis and medialis) tend to occur
almost simultaneously (Kelley et al. 1976, Millington et al. 1992, Khemlani et al.
1998), demonstrating peak activity around TO which diminishes as standing is
reached. Variable patterns of EMG activity have been found in gastrocnemius
and soleus muscles, probably reflecting their additional role in balancing the
body mass throughout the action (e.g. Khemlani et al. 1998).
Rectus femoris is not activated early enough to be associated with initial hip flex
ion (Kelley et al. 1976, Millington et al. 1992, Khemlani et al. 1998) and iliop
soas, by its action on the pelvis and thigh, would be the major initiator. The
simultaneous onsets of biarticular rectus femoris and biceps femoris (Khemlani
et al. 1998) may reflect their contributions to controlling hip flexion and therefore
body movement forward, with rectus femoris contributing to flexion and biceps
femoris exerting a braking moment at the hip that serves to decelerate upper
body movement prior to the onset of extension. Both biceps femoris and rectus
femoris continue to contract after thighs-off. The co-contraction of two theoret
ically antagonistic muscles such as these is efficient, each muscle reinforcing the
other as they combine to extend hip and knee (Roebroeck et al. 1994).
Sitting down
On observation, STS and SIT appear fundamentally the same action only in
reverse. Certainly, angular displacements at hip, knee and ankle joints are simi
lar (Kralj et al. 1990). Mechanically, however, the two actions are different
and independent. SIT involves flexion at hips, knees and ankles controlled by
extensor moments of force at these joints, i.e. the lower limb extensor muscles
spanning the three joints are generating and sustaining lengthening (eccentric)
contractions throughout the movement. The body mass moves back as well as
down, the upper body flexing forward. SIT lacks the propulsive component
that is so critical in STS. Movement duration tends to be longer, a precaution
due to lack of visual information and uncertainty about seat location (Kralj
et al. 1990). Since balance mechanisms differ in this action compared to STS, it
is critical that it is trained specifically.
Difficulty standing up from a seated pOSitIOn is common in elderly people.
Undoubtedly factors such as joint pain, muscle stiffness, muscle weakness and
decreased range of motion arising from pathology and physical inactivity can
affect the performance of both STS and SIT in this age group. However, healthy
elderly subjects standing up under constrained laboratory conditions without
arm use have demonstrated motor performance similar to young adults in terms
of angular displacements, timing of joint onsets and moments of force normal
ized to subject's body mass (Ikeda et al. 1991, Saravanamuthu and Shepherd
1994). Although increased movement duration has been reported in older
individuals, elderly subjects are able to increase speed of rising when asked to
stand up as fast as possible (Vander Linden et al. 1994).
A study of STS examined a group of able-bodied elderly people, a group of
elderly who could not stand up without using their arms and a group of young
able-bodied adults (Schultz et al. 1992). Medical histories and physical findings
were essentially normal in the young group but were important in differenti
ating the elderly groups. Every subject in the group who had to use their arms
had some degree of lower limb muscle weakness and joint deformity (primarily
due to osteoarthritis), and problems such as dizziness and decreased vision. Move
ment time was similar in the young and able-bodied elderly groups but signifi
cantly longer in the group who used their arms. Both older groups had
significantly increased angular displacements at hip and knee joints compared to
the young, but only moderate differences in joint moments of force at thighs
off. The larger joint displacements resulted in a more anterior position of the
centre of body mass at thighs-off. This strategy may provide greater stability,
since the centre of body mass at thighs-off would be within the foot support
area. The results suggest that hands may be used to gain more postural stabil
ity, particularly at TO, rather than to reduce joint moments. However, using
the hands at TO may also be a strategy to decrease force requirements at the
knee in individuals with substantial joint pain.
Research findings
Despite the importance of STS as a precursor to getting out of bed and walking
there has been little investigation of this fundamental action following stroke.
In general, movement time and vertical ground reaction forces (GRF) under
each foot are the variables that have been tested. Although different method
ologies and protocols have been used, there are two consistent findings:
Stroke patients take significantly longer to stand up than age-matched able
bodied subjects (Yoshida et al. 1983, Engardt and Olsson 1992, Hesse et al.
1994). Sitting down is also slower (Yoshida et al. 1983, Engardt and Olsson
Amplitude of vertical GRF under the paretic lower limb is markedly
decreased compared with the non-paretic lower limb (Engardt and Olsson
1992, Hesse et al. 1994, Fowler and Carr 1996).
Able-bodied subjects of different ages stand up at their preferred speed in
approximately 1.6 s, with the pre-extension phase occupying approximately
30% and the extension phase 70% of the total movement time. Differences in
defining movement onset and movement end make it difficult to generalize.
However, in one study, stroke subjects tended to stand up very slowly, taking
on average 2.3 s to complete the extension phase (Ada and Westwood 1992).
Associated with this latter finding were angular displacements at hip and knee
that occurred at low velocities with no discernible peak. After specific training
.------rypical vertical ground reaction force (GRF) profile during STS performed by a
stroke patient. Lower trace shows the hemiparetic limb.
300 I
200 i
0 1 2 3 4 5
Time (s)
of STS, however, duration of the extension phase decreased to 1.5 s. Knee and
hip velocities showed a definite peak, the plots assuming a clear bell shape typ
ical of well-coordinated movement, i.e. as performance improved, coordin
ation between hips and knees was regained. The change from independent
action at each joint to a more normal coupling of the joints demonstrates the
development of the basic synergy required for unaided standing up. The coord
ination changes shown in this study are similar to those of able-bodied people
developing skill (Ada et al. 1993). Another study also reported an increase in
speed after training of the action (Engardt et al. 1993). Although both studies
involved small samples, these results illustrate the potential of stroke patients
to improve with appropriate and intensive training (Table 4.1).
Vertical ground reaction force graphs typically show either no discernible
peak at thighs-off for the paretic leg or peak force occurring later in the action
(Fig. 4.8). Fluctuations in vertical force both during the action and in quiet
standing (Yoshida et al. 1983, Fowler and Carr 1996) can indicate difficulty in
balancing the body mass over both feet.
Observational analysis
The therapist observes the performance of STS and SIT or the patient's attempts
at these actions, comparing performance with a list of key components derived
from biomechanical data reported consistently in able-bodied subjects across
TABLE 4.f Sit-to-stand: clinical outcome studies
Reference Subjects Methods Duration Results
Ada and Westwood 1992
Engardt et al. 1993
Fowler and Carr 1996
8 Ss
Early post-stroke
Mean age 65 8.9 yrs
40 Ss
<3 months post-stroke
Mean age
E: 64.6 6.7 yrs
C: 65.1 9.0 yrs
12 Ss
<3 months post-stroke
Mean age 63.5 (43-79) yrs
Pre-test, post-test design
Task-specific training of STS
Pre-test, post-test design
Randomly assigned to two groups:
C: Task-specific training of STS/SIT
E: Task-specific training of STS/SIT
with auditory feedback about force
produced by paretic leg
Pre-test, post-test design
Randomly assigned to two groups:
1: Task-specific training of STS
2: Task-specific training + auditory
feedback about force produced by
paretic leg
4 weeks
5 days/week, 10-20 mins
6 weeks
5 days/week, 15 mins
3 weeks
25 repetitions twice daily
Significant decrease in movement
duration (p < 0.005), increased
amplitude of peak angular
velocities (p < 0.01). All
subjects reached 6 on Motor
Assessment Scale (Item 3)
Coordination of hip and knee
angular velocity more
STS - significant increase in
force through paretic leg in
both groups (p < 0.01)
SIT - significant increase in force
produced through paretic leg
in E group only (p < 0.001).
All subjects but one produced
peak force ~ 50% body weight
through paretic leg at
end of study
__ (Continued)
Reference Subjects Methods Duration Results
Hesse et al. 1998
Cheng et al. 2001
35 Ss
3-23 weeks post-stroke
Mean age 64.8 (59-79) yrs
Note: all subjects could
stand up independently
with minimum help
54 Ss
<4 months post-stroke
Mean age 62.7
Pre-test, post-test design
Bobath techniques (tone inhibiting,
trunk control in lying and sitting,
pelvic tilting, weight shift in standing,
facilitated walking)
Randomly assigned to 2 groups:
C: (neuromuscular facilitation, FES,
mat exercises, therapeutic exercises)
E: (symmetrical standing and
repetitive STS training with visual
and auditory feedback)
4 weeks
4 days/week, 45 mins
C: not specified
E: 3 weeks 5 days/week,
50 mins
STS - no change in movement
duration, horizontal and
vertical GRFs
Gait - no change in speed,
cadence, stride length.
Rivermead Motor Score,
Ashworth score, Motricity
Index showed little or no
relevant change
Follow-up at 6 months
C: no significant differences
between initial and follow-up
tests of STS and SIT
E: significant differences
between initial and follow-up
tests - decrease in duration of
STS (p < 0.001) and SIT
(p < 0.01); decrease in LlR
vertical GRF - STS (p < 0.005),
SIT (p < 0.001); decrease in
COP sway STS (x: p < 0.01,
y: p < 0.05), SIT
(x: p < 0.005); increase in rate of
rise in force in STS (p < 0.001)
Significantly fewer falls in
E (p < 0.05)
C, control group; COP, centre of pressure; E, experimental group; FES, functional electrical stimulation; GRF, ground reaction force; SIT, sitting down; Ss, subjects: STS, sit-to-stand.
different age groups. The following components appear to be critical to effective
initial foot placement backward (approximately 10 cm from a vertical
line dropped from the knee joint, approximately 75 ankle dorsiflexion)
flexion of the extended trunk at the hips* to move the body mass
forward, together with dorsiflexion at the ankles
knee, hip and ankle extension.
The above list comprises sagittal plane angular displacements at hip, knee and
ankle which are easily observable by the clinician. Also observable are paths of
the shoulder and knee as the individual performs the action. The shoulder moves
along a smooth, continuous curvilinear path forward and upward. As the hips
flex, the knees move forward along a linear path for a few centimetres, move
ment occurring at the ankles.
These components provide a structure or a model of the action with which to
compare the patient's performance and determine major adaptations and pos
sible causes. Although it is not possible to observe the underlying momentum
and force characteristics of the action it is possible to make inferences from
observational analysis about these characteristics. Knowledge of three mechan
ical requirements for effective standing up is necessary as a reference:
generation of sufficient speed and therefore horizontal and vertical
momentum to propel the body forward and upward over the feet
generation and sustaining of lower limb joint forces to support and raise
the body mass into standing
postural stability at thighs-off, controlling the centre of body mass in
relation to foot support area.
Some common motor problems evident on observational analysis and possible
explanations are described below. They include the following.
Inability to stand up independently
- lower limb muscle weakness and lack of coordination.
Absent/insufficient foot placement backward (flex knee and dorsiflex
ankle) (Fig. 4.9a)
- weakness of hamstrings and ankle dorsiflexors
- decreased extensibility of soleus.
Decreased projection of body mass forward in pre-extension phase (flex
hips and dorsiflex ankles) (Fig. 4.9b)
inadequate foot placement backward
inability to stabilize the paretic lower limb, in particular the shank and
foot. Normally the lower limbs are used actively for balance and
propulsion in the pre-extension phase
flexion of spine instead of at hips (Fig. 4.9c)
moving very slowly
"Although the pelvis is the segment linking the lower limbs to the spine, during STS the pelvis
and spine act together as a virtual segment as the upper body pivots forward at the hips.
Minimal movement in the spine and between the spine and the pelvis has been reported.
Although sometimes referred to in the clinic as forward pelvic tilt, the movement is actually a
rotation of the pelvis-spine unit by flexion at the hips.
144 Training guidelines
FIGURE 4.9 (a) Failure to place paretic leg back. Adaptations: weight concentrated on non-paretic
R leg; increased distance between feet. (b) Failure to move body mass forward over
feet (insufficient hip flexion and ankle dorsiflexion). Note adaptive use of arms.
(c) Failure to move body mass forward (flexing lumbar spine instead of hips). Note
adaptive use of arms. (d) In sitting down, body mass has not stayed forward over the
feet as hips, knees and ankles flex. As a result she has lost balance backward and falls
to the seat. (e) Feet have not been moved backward and arms are swung forward to
aid propulsion forward and up.
(a) (b) (c)
(e) (d)
Standing up and sitting down 145
- fear of falling forward
- lack of vigour.
Decreased control of body mass sitting down (Fig. 4.9d)
- weakness of eccentric control of lower limb extensor muscles.
Decreased stability, particularly at TO when horizontal momentum of the
body mass must be controlled and flexion of upper body at hips changes to
weakness of lower limb muscles and slowed reaction times (muscles
that link the shank and foot are particularly critical to balancing the
body mass as it pivots over the feet)
lack of intersegmental coordination and poor timing.
There are several predictable adaptations to muscle weakness, lack of segmen
tal coordination and decreased stability. They include:
weight borne principally on stronger and more easily controlled leg
body mass can be observed to shift laterally toward non-paretic leg at TO
positioning of foot of stronger leg behind foot of weaker leg (the most
posterior foot bears most weight)
use of hands for balance or support
arms swung forward to aid horizontal and vertical propulsion of the body
mass (Fig. 4.ge)
distance between feet increased (Fig. 4.9a)
action is performed slowly with a pause between pre-extension and
extension phases.
Training of STS and SIT following stroke is critical for the regaining of
independence and mobility. Inability to perform these actions effectively and
efficiently predisposes the individual to dependence on others for mobility
and to an increasingly sedentary lifestyle. It has also been reported that
being assisted to stand up by another person can be a common cause of
damage to the soft tissues around the shoulder complex and the
development of pain (Wanklyn et al. 1996).
Immediately following stroke, patients naturally adapt their performance
as they attempt to become independent and learn to manage daily
activities. The principal adaptation is to bear weight mainly through the
stronger leg and use the non-paretic upper limb. If these adaptations
persist, the patient may learn not to use (i.e. neglect) the affected limbs.
It appears from stroke rehabilitation texts that therapy methods may
reinforce this adaptation. The emphasis in clinical practice may not be
on training standing up and sitting down and on strengthening lower
limb muscles, but on teaching what is to the patient the novel task of
'transferring' from one seat to another. There are several variations of the
transfer (low transfers, pivot transfers, STS transfers). All involve the
patient being taught to bear weight on the stronger leg and arm, with
146 Training gUidelines
minimum support from the affected leg, in order to move from one sitting
surface to another. Although it may be a practical necessity for patients to
be able to move from, say, wheelchair to toilet, this manoeuvre can be
practised by the patient in the appropriate environment rather than taught
as a general strategy.
Every effort should be made to train STS and SIT intensively, since real
world independence requires not the ability to transfer from one surface
to another by pivoting on the non-paretic limb but the ability to stand up
and sit down. There is some evidence that repetitive training of STS and
SIT has some generalizability. The more the paretic limb is loaded during
STS and SIT the better the individual performs in other activities of daily
living as measured on the Barthel Index and the Fugl-Meyer Scale
(Engardt et al. 1993).
Physiotherapy intervention has the potential to train most patients to gain
independence in these actions. Although muscle weakness is a limiting factor
immediately after stroke, there are several key mechanical factors arising
from biomechanical research which can be utilized to make it possible for
even a frail patient to practise the action:
initial foot position at approximately 75
ankle dorsiflexion as this
requires less effort
vary seat height depending on strength of lower limb extensors
active hip flexion initiated with the trunk vertical (90
to horizontal) to
optimize horizontal momentum of the body mass
no pause between pre-extension and extension phases
speeding up the action.
For training to be effective, given the impairments following stroke, it is also
necessary to:
prevent adaptive soft tissue shortening
increase lower limb muscle activation, strength and coordination.
The following guidelines provide what can be considered a critical
pathway for optimizing STS/SIT for most individuals following stroke.
The guidelines are individualized according to each patient's impairments,
level of functional ability and needs, and are set down with no intention
of a prescribed sequence. They address the essential requirements for
effective performance, synthesized from research.
Standing up and sitting down practice
Standing up
Start with upper body vertical and feet placed back. Patient swings upper
body forward at the hips and stands up (Fig. 4.10).
Patient sits on a firm flat surface.
No arm rests.
Standing up and sitting down 147
FIGURE 4.10 Practice of standing up and sitting down. (a) Therapist stabilizes paretic foot on the
floor by pushing down along the line of the shank. (b) As she stands up therapist
moves her body mass laterally to increase loading of the L leg. (c) Lateral view
illustrates how the therapist can give guidance for relative intersegmental alignment.
It is important that the patient does not extend her upper body at the hips before
the body mass is for enough forward.
(a) (b) (c)
Feet flat on floor (vary thigh support so ankles are dorsiflexed to 75).
No flexion within upper body throughout action, eyes on target.
Choose a seat height to encourage weightbearing through the affected
lower limb and discourage adaptive loading of the non-paretic limb
(Fig. 4.11).
Directing vIsion and attention toward a target during STS/SIT assists
appropriate body alignment by affecting head position and sense of verti
cality. Target is placed 2-3 m in front at eye level.
If necessary, therapist assists foot placement so the action is practised
from an optimal starting position. Voluntary activation of hamstrings and
ankle dorsiflexors for foot placement are trained specifically in Sitting.
For patients with severe muscle weakness, therapist stabilizes the foot
on the floor and encourages weightbearing through the affected leg by
pushing down and back along the line of the shank. This ensures that
quadriceps activity moves the thigh on the shank to extend the knee
148 Training guidelines
FIGURE 4.11 (a) Although she can stand up independently, she loads the L leg more than the R. Note
that she has moved the foot of her non-paretic L leg further back just before thighs-off.
(b) It is better for her to practise repetitively from a higher seat. Left: on her early
attempts she moved the L foot behind the paretic foot. Right: the physiotherapist
suggests the R foot is placed further back to force more loading of the paretic R leg.
Standing up and sitting down 149
FIGURE 4.12 (a) Vertical ground reaction forces during unconstrained standing up. (b) The effect of
the therapist stabilizing the affected foot on the floor with some external pressure
down and back along the line of the shank. Note the redistribution of the vertical
ground reaction forces through the two legs.
-05 -0.4 -03 -02 -01 0.0 0.1 02 0.3 0.4 05
Time (seconds from thighs-off)
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
Time (seconds from thighs-off)
rather than the foot sliding forward. Guiding the knee forward along a
horizontal path may help the patient move the body mass forward over
the feet.
Do not block forward movement of the body mass by standing too close
to the patient, i.e. blocking shoulders or knees from moving forward.
Give instructions that identify body part to be moved, e.g. 'Swing your
shoulders forward and stand up'. The instruction 'Lean forward' is not
an appropriate cue as it does not identify what body part is to be moved
or take account of the generation of horizontal momentum.
Discourage use of hands.
Discourage holding paretic hand in other hand. This does not increase
weightbearing on paretic lower limb or symmetry (Seelen et al. 1995).
Standing up and sitting down 149
FIGURE 4.12 (0) Vertical ground reaction forces during unconstrained standing up. (b) The effect of
the therapist stabilizing the affected foot on the floor with some external pressure
down and bock along the line of the shonk. Note the redistribution of the vertical
ground reaction forces through the two legs.
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
(a) Time (seconds from thighs-off)
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
(b) Time (seconds from thighs-off)
rather than the foot sliding forward. Guiding the knee forward along a
horizontal path may help the patient move the body mass forward over
the feet.
Do not block forward movement of the body mass by standing too close
to the patient, i.e. blocking shoulders or knees from moving forward.
Give instructions that identify body part to be moved, e.g. 'Swing your
shoulders forward and stand up'. The instruction 'Lean forward' is not
an appropriate cue as it does not identify what body part is to be moved
or take account of the generation of horizontal momentum.
Discourage use of hands.
Discourage holding paretic hand in other hand. This does not increase
weightbearing on paretic lower limb or symmetry (Seelen et al. 1995).
~ 5
LL 4
(ii 3
to: 2
- Unaffected
1SO Training guidelines
If patient is flexing within the spine and not at the hips, direct attention
(vision) toward the target.
Final standing alignment is with hips and knees extended, angles at
approximately 0. Avoid assisted extension of the knee while the hips
are still extending or patient will overbalance backward.
Seating. Chairs provided for stroke patients frequently enforce passivity.
In choosing a seat, a compromise should be achieved between comfort and
ease of standing up. Patients early after stroke and those who are weak are
not able to stand up unaided without the provision of such a chair. If there
is no supportive muscle activity around the shoulder, the arm should be
supported when the patient is sitting (see Fig. 5.23).
Sitting down
Patient flexes hips, knees and ankles to lower body mass toward the seat.
Upper body flexes forward at the hips as body mass is lowered.
Weight remains supported over the feet. Nearing the seat, body mass is
moved back to enable seat contact.
Assist initiation of knee flexion by moving the knee forward.
Stabilize shank and foot to assist weightbearing on affected leg.
Specific training of SIT with weight through the affected leg appears neces
sary to improve performance (Engardt et al. 1993).
Soft tissue stretching
Extensibility of soleus muscle is critical to foot placement backward and to
weightbearing through the affected leg. Preserving functional muscle length
involves both passive and active stretching.
A brief passive stretch is given immediately before exercises to decrease
muscle stiffness. Active stretching occurs during practice of STS and SIT and
the weightbearing exercises below if the paretic limb is loaded and ankles
are initially dorsiflexed.
Soleus: stretch in sitting (Fig. 4. 13).
Hold stretch for approximately 20 s, relax, repeat 4-5 times. Patients
can learn to do this themselves.
Patients with severe muscle weakness who are physically inactive require
periods of prolonged passive stretching during the day.
Soleus: stretch in sitting (Fig. 4.14).
Gastrocnemius-soleus: prolonged stretch in standing (see Fig. 3. 18c).
Standing up and sitting down 151
FIGURE 4.13 Patient stretches her L soleus for 20 s prior to practice of STS and SIT.
FIGURE 4.14 Sustained passive stretch to soleus. (Conceived and developed by K Schurr and
JNugent, Physiotherapy Department, Bankstown-Lidcombe Hospital, Sydney.)
152 Training gUidelines
FIGURE 4.15 A roller skate reduces the friction between the shoe and floor while she practises
activating hamstrings and muscles crossing the ankle to move her foot back. She aims
to move her toes behind a line on the floor (concealed under rear wheel).
Hold prolonged stretch for 30 min.
Eliciting muscle activity
Initial foot placement backward is critical to the ease of STS since it
minimizes the distance the body mass is moved forward over the feet.
Specific training to activate hamstring and anterior tibial muscles may be
necessary. The therapist can palpate hamstrings to ensure they are active
or use an EMG monitor. Activity of tibialis anterior is easily observable.
Foot placement backward
Patient sits with feet flat on floor and slides foot backward and forward.
Decreasing resistance using a slippery surface or roller skate may
encourage this action (Fig. 4.15).
Encourage patient to lift thigh slightly to reduce friction under the foot.
Once patient voluntarily activates these muscles and gets the idea of the
movement, encourage an increase in speed.
Standing up and sitting down 153
FIGURE 4.16 Getting the idea of moving the body mass forward and backward over the feet by
f'exing at the hips and dorsi f'exing at the ankles. This is useful when the patient
is fearful of moving forward. Paretic foot may need to be stabilized.
Incorporate into STS practice, Le. expect the patient to move the foot
back without assistance.
Electrical stimulation to ankle dorsiflexors in sitting may assist the
Moving the upper body forward and backward at the hips
Fear of falling may be a major barrier to moving the body mass forward in
some patients early after stroke. This action may require some practice
briefly at the start of rehabilitation to increase the patient's confidence.
Patient sits with arms on a table at approximately shoulder height, trunk
and head erect. She flexes the upper body forward at the hips by sliding her
hands to the edge of the table and back to the erect position (Fig. 4. 16).
Maintenance of erect trunk and head is encouraged by having the patient
look ahead at target. The far edge of the table needs to be a distance of
at least 1.5 times arm length to enable body mass to move forward over
the feet.
154 Training guidelines
Assist movement of paretic arm if necessary.
Stabilize the foot on the floor to encourage generation of ground reaction
forces and activity in anterior tibial muscles as the person moves forward.
Encourage patient to take weight through the feet and to speed up the
Incorporate into STS practice.
Strength training
Lack of strength, particularly in the lower limb extensor muscles, is a major
limiting factor in the ability to stand up and sit down, together with lack of
intersegmental coordination and postural instability. Functional strength
training includes repetitive practice of the action itself.
STS and SIT with progressive body weight resistance
Practice of STS/SIT with maximum repetitions and against an increasing
load (body weight) can increase and maintain strength in the lower limb
extensor muscles (Kotake et al. 1993, Carr and Shepherd 1998).
Progressive strength training involves lowering seat height and increasing
number of repetitions. Patient is encouraged to do a maximum number of
cycles of STS/SIT for a selected seat height, up to 10, repeated three times.
When this can be achieved, the seat is lowered. Seat height selection is
determined by observation of the ability to bear weight relatively evenly
through both lower limbs.
Many repetitions without pause are necessary to improve strength and
control, so a target number should be set and repetitions counted
(Fig. 4.17a).
Placing foot of weaker limb behind other foot forces weightbearing
through the weaker leg (vertical GRF are greater under the backward
placed foot, particularly at TO)
Lower limb muscle strength enhances the generation and control of
momentum and dynamic stability, ultimately influencing the ability to
stand up and sit down (Scarborough et al. 1999).
As strength increases, encourage an increase in speed. Reduce upper
limb assistance by folding arms (Fig. 4.17b).
This is a suitable home exercise to be carried out long-term, particularly
by elderly individuals, to preserve lower limb muscle strength and func
tional ability.
The patient may also benefit from strengthening exercises that have similar
dynamics to STS/SIT, i.e. flexion and extension of the lower limb joints to
raise and lower the body mass over the feet:
step-ups (see Fig. 3.22)
heels raise and lower (see Fig. 3.24).
Standing up and sitting down 155
FIGURE 4.17 (a) Partially supervised repetitive practice ofSTS and SIT. She is getting feedback
about loading the limb from a limb load monitor and is recording the number of
repetitions on a counter. (b) Repetitive practice with arms folded increases lower
limb extensor force requirements. * (*Courtesy of K Schurr and S Dorsch,
Physiotherapy Department, Bankstown-Lidcombe Hospital, Sydney.)
(a) (b)
The above exercises can be practised in group circuit training with minimal
Maximizing skill
In the initial stages of rehabilitation the patient and therapist focus on
developing the basic movement coordination, i.e. the essential spatial and
temporal components of STS/SIT, using many repetitions. Although
repetitive practice plays an important role in tuning motor activity to
become effective and more reliable (Gallistel 1980), variable practice - what
Bernstein (1967) has described as 'repetition without repetition' - is
necessary to develop skill. There is a great improvement in morale and
lifestyle when a person can stand up from a chair unaided. However, for
156 Training guidelines I
FIGURE 4.18 Training to increase stability and skill: balancing the mugs on the tray changes the
focus of attention.
flexibility of performance, practice has to provide the opportunity for the
individual to learn how to adapt performance to different contexts.
Training for skill therefore includes practice of STS/SIT under different
regulatory conditions:
Holding a glass of water, tray with utensils (Fig. 4.18), objects of different
sizes and weights; carrying on a conversation (to decrease conscious
attention to performance); varying speed.
Performing action sequences: STS ----j walk to the right or to the left, to
another chair ----j SIT ----j STS ----j walk back ----j SIT.
Stopping on request (immediately after thighs-off or before thighs-on)
without losing balance.
STS/SIT from different types of seat.
The most commonly reported motor problems in STS and SIT are moving slowly
compared with age-matched able-bodied subjects, decreased weightbearing on
Standing up and sitting down 157
the affected lower limb and poor balance. Since observational analysis gives
uncertain results, valid and reliable tests are necessary to document progress
and the effect of intervention. The following tests have been found to be reli
able if performed under standardized conditions.
Functional tests
These include:
Motor Assessment Scale: Standing up item (Carr et al. 1985)
Timed 'Up and Go' Test (Podsialo and Richardson 1991)
Quadriceps strength using a hand-held dynamometer.
Timed STS tests may also provide an indirect measure of knee extensor muscle
strength (Fiatarone et al. 1990, Bohannon et al. 1995). Bohannon and col
leagues reported a curvilinear relationship between isometric knee extension
strength measured with a dynamometer and the number of STS/SIT performed
in lOs. The number of repetitions also correlated with maximum gait speed.
Movement duration can be tested simply, using a visual estimate of the start
and end of STS and SIT. Starting and finishing positions for both must be stan
dardized. Movement is said to begin when the subject's head starts to move
anteriorly and ends when no further displacement of the pelvis occurs. SIT
begins when the head starts to move and ends when the pelvis rests on the seat.
The technique on which this is based was found to be reliable when observed
time values were compared with data from a motion analysis system (Engardt
and Olsson 1992). The tester stands at the side of the subject with a stopwatch.
Biomechanical tests
Variables that can be tested include:
movement duration
magnitude and timing of vertical ground reaction force
coordination of angular displacement and velocity - phase-plane plots
(Cahill et al. 1999).
Augmented feedback
Results of research on the efficacy of providing auditory feedback, whether
using a specially designed vertical ground reaction force feed back device
(Engardt et al. 1993) or a Limb Load Monitor" (Fowler and Carr 1996), are
"Electronic Quantification, Model 101-1.
158 Training guidelines
equivocal. It may be that focusing on making the auditory signal can become
the goal of the action rather than loading the limb and developing an internal
representation of the action. In addition, an auditory signal about force
already produced may come too late to allow modification of the action
(Fowler and Carr 1996). A visual template of the specific force-time curve
required may, therefore, be a more beneficial form of feedback than informa
tion about peak force itself. However, as with the use of any external feed
back, the patient must be able to move on to use naturally occurring feedback
about weight distribution.
Reaching and
echanical description
ing to grasp
very of upper limb
.on after stroke
of motor
Eliciting muscle activity
Reaching and balancing practice
Manipulation and dexterity
I practice
pain prevention
pecific methods
t therapy'
omputerized training
Much of what we do with our arms is related to carrying out complex tasks
involving one or both hands. We move our arms in order to place our hands at
the appropriate place for manipulation in the working environment and to trans
port objects from one place to another. The distance we can reach is a function of
arm length. If the object to be reached is beyond arm's length, protraction of the
shoulder girdle and movement of the body mass forward at the hips and ankles
extends the reachable distance. If the object is out of reach in standing a step is
taken. The muscle forces produced and the timing and sequencing of joint move
ments involved in a specific action are a function of the task being performed, the
object, the individual's position relative to the object, and the constraints of the
environment. The way we move is controlled by the context; for example, both
the shape of an object and what is to be done with it determine the type of grasp.
The location of glass relative to mouth determines how the hand moves through
space, and the amount of rotation of shoulder and forearm. Bernstein observed
that dexterity is never present in the motor act itself but only in the interaction of
the act with the changing environment (cited in Latash and Latash 1994).
The large variety of motor actions performed by the arm and hand illustrate
two fundamental problems in motor control raised originally by Bernstein
(1967): the many degrees of freedom available in the neuromusculoskeletal
system and the need for context-specific or context-conditioned variability.
Here also is a dilemma for stroke rehabilitation. As a result of the lesion there
are a very large number of actions, ranging from simple to complex, which the
patient must relearn in the contexts of independent daily life. Most important
therefore is functional training in which the person learns to optimize motor
control. Training is designed to help the patient regain the ability to harness
the degrees of freedom available so the limb functions as a coordinated unit in
functional actions with many different goals.
Skilled motor actions are characterized by patterns of segmental movement
which best address the spatiotemporal demands of the action. Although theoret
ically there are many movement options due to the flexihility of the neuromuscu
loskeletal system (the degrees of freedom), the most parsimonious (cost-efficient)
options are used (Kelso et al. 1994). Normally, the control system utilizes 'sim
plifying' strategies made possible by anatomical configurations, neuronal connec
tions and biomechanical characteristics of linked segments. For example, the arm
and hand appear to function as a single coordinated unit in reaching and manipu
lation as do the fingers in grasping; in bimanual actions the two limbs may initi
ate and end motion simultaneously despite differing constraints for each arm.
The concept of simplification also enables the development of exercises, the
effects of which are likely to transfer to improved performance of a number of
different tasks.
Although there are a large number of actions that we carry out every day, they
fall into different categories. Some have high externally imposed temporal
demands, such as catching a ball; others have high precision demands, such as
picking up a pin or keyboard typing. Many tasks involve both hands if they
are to be carried out easily and effectively. In bimanual actions such as opening
a can of drink the two limbs are constrained to function as a coordinated unit.
This means it is unlikely that individuals with predominantly unilateral impair
ments will function effectively if they are not specifically trained in bimanual
actions. This creates another dilemma in planning intervention. Exercises are
necessary to activate and train weak muscles of the affected limb, yet emphasis
must also be on bimanual practice. For many patients the optimal intervention
for avoiding learned non-use and for training functional use of the affected
limb involves intensive exercise and training of the affected limb, in some cases
forcing use by constraint of the non-paretic limh. Daily training sessions have
to be planned, however, to incorporate intensive training of bimanual tasks.
Most actions specifically involving the upper limbs also involve the whole body
due to the dynamics of the segmental linkage. The arms also playa part in bal
ancing mechanisms (Gentile 1987). When the centre of body mass moves too
close to the limits of stability they may playa stabilizing and supportive func
tion. If balance is lost and the person falls, the arms can playa supportive role
or, at the very least, serve to break the fall.
The function of the hands is sensory as well as motor (Hogan and Winters
1990). Tactile receptors in the palm convey critical information for the identifi
cation of objects and their properties. Tactile information is obtained from the
feel of the object in the hand, the recognition of its nature (size, shape, form in
three dimensions, composltlon and texture) and its posItIOn in the hand.
Proprioceptive information involves awareness of the relationship of parts of
the hand to each other, and their position in space. A significant aspect of sens
ory discrimination for manipulation is the appreciation of the compressibi
lity of an object - a combination of dynamaesthesia (the awareness of force
applied in a motor act), kinaesthesia (appreciation of change in position of fin
gers) and counter-pressure of fingertips (Roland 1973).
Vision plays a critical role in reaching and interacting with objects. The
dynamic linkage between eye, head and upper limb movements enables skilled
hand use. Information of importance to understanding motor control comes
from the work of Gibson (1977), who used the term 'affordances' to refer to
the properties of the environment to which we are attuned. When we ask a
person to look at what he or she is doing, we expect the person to pick up
information about the object and the environment (the affordances or possibil
ities offered for interaction) that will assist in task performance.
Major prerequisites for hand use are therefore the ability:
to move the hand to the scene of the action
to look at and pay attention to both object and environment
to make the postural adjustments which occur with arm movement
to utilize somatosensory information.
Many early scientific investigations of upper limb function examined arm and
hand movements in simple reaching actions or manipulation under relatively
constrained conditions. More recently, researchers have studied the inter
actions between reaching and manipulation in real-life functional tasks.
Reaching to grasp
Reaching to grasp an object can be divided into two components: transporta
tion and manipulation (Jeannerod 1981). The transport component is per
formed relatively ballistically. The manipulation component, however, requires
visual feedback to ensure an accurate approach with appropriately sized grasp
aperture. Studies of the relationship between these two phases (Marteniuk
et al. 1990, Hoff and Arbib 1993, Rosenbaum et al. 1999) show that the arm
and hand function as one unit during the reach, with the hand starting to open
around the time the reaching action begins. Final adjustment to the grasp aper
ture occurs at the end of the reach just prior to grasp (Fig. 5.1). The magnitude
of grasp aperture reflects the size of the object, with an extra few centimetres
as a margin for error. Grasp aperture is also related to the speed of the move
ment, wider apertures being observed with faster movements (Wing et al.
1986). Stabilization of the thumb in abduction may provide a focus for moni
toring the relationship between grasp aperture and object size (Fig. 5.2). Just
before the object is grasped the aperture decreases and aperture adjustment
. ...........
Marke( over thumb
Approximate points of
contact on thumb and
/;. finger pads

Marker over wrist
Distal interphalangeal
joint of index finger ----:
Marker over
joint of index
A cinematographic study by Jeannerod of reaching toward an object. Dots represent
successive positions of the hand every lOms. Note the hand slows on approach to the
object. Lines represent the size of the grasp aperture every 40 ms. Note that aperture
formation starts soon after movement onset, reaches a maximum of 3 cm, then
decreases until it is appropriate for the size of the object. (We thank the International
Association for the Study of Attention and Performance for permission to reprint from
Long and Baddeley 1981, as cited in Jeannerod 1981.)
aperture opens and closes by movement of the fingers to and from the abducted
thumb. Figure shows the marker positions used in the kinematic investigation.
(Reproduced from Wing and Fraser 1983, with permission.)
appears to be a function of finger movement to the already abducted thumb
(Wing and Fraser 1983).
The spatiotemporal organization of these transportation and manipulation
components varies according to whether one or both hands are used (Castiello
et al. 1992). It takes into account the shape and other physical characteristics
of the object, including its fragility, and what is to be done with it (Iberall et al.
1986, Rosenbaum et al. 1992, van Vliet 1993, Smeets and Brenner 1999).
The organization of transport and manipulation components differs according
to whether subjects reach to grasp with the whole hand or with a precision
grasp. In opening a can, for example, one hand reaches to hold the can and
stabilize it, the other reaches to grasp and pull the ring with thumb and index
finger. A study of this action in able-bodied subjects (Castiello et al. 1993)
showed that both hands started and finished the movement together, although
the kinematic details varied for each hand (Fig. 5.3). In particular it was noted
that the precision hand reached a smaller value of peak velocity earlier and
started to decelerate earlier than the other hand. The longer deceleration time
illustrates the relative complexity of the ring pulling task compared to the
holding task. A similar relationship between the two limbs was found in
another task that involved opening a drawer with one hand and taking out a
small rod from the drawer with the other (Wiesendanger et al. 1996).
Cooperation between the two hands and the predictive nature of some tasks are
also shown by an experiment in which the subject dropped a ball into a cup
held in the other hand (Johansson and Westling 1988). The grip force of the
cup hand was found to increase in anticipation of the ball's impact, i.e. before
the ball hit the cup. Findings from such studies confirm the view that a patient
with a unilateral lesion is unlikely to regain the ability to perform bimanual
actions unless such actions are practised. It is only by practising tasks with both
hands that interlimb timing parameters can be regained. In addition, bimanual
exercise addresses the issue of poor dexterity (clumsiness) in non-paretic hand
use, which may be present due to the effects of the brain lesion on ipsilateral
The upper limbs are also involved in throwing and striking actions. Most of
these actions are characterized by sequential movement of the segments within
the upper limb linkage, proximal segments moving first. This sequencing imparts
maximum speed effects on the distal segment and therefore the ball (Putnam
Any movement of the body mass toward the limits of stability, such as reach
ing beyond arm's length, has the potential to destabilize the individual. In both
sitting and standing, movement of the body over the base of support is nor
mally countered both before and during the reach by postural leg muscle activ
ity which is linked to the arm movement (Ch. 2).
There are a number of degrees of freedom to be harnessed in actions involving
the arm in sitting and standing. Studies of the relative contributions of trunk
and arm movement in reaching beyond arm's length in sitting have shown
- Right hand:
whole hand prehension
Left hand: precision grip
800 0
Time (ms)
Left hand:
, whole hand prehension
Right hand: precision grip
Left hand:
whole hand prehension
variability in performance according to task and context, but consistency in
performance between subjects (Kaminski et al. 1995, Dean et al. 1999a,b,
Seidler and Stelmach 2000). The task itself affects temporal aspects of coord
ination, with simple tasks (pressing a switch) being carried out faster than more
complex ones (reaching to pick up a full glass of water) (Dean et al. 1999a).
On the other hand, distance to be reached affects spatial and temporal coord
ination between trunk and arm (Saling et al. 1996, Dean et al. 1999a), with
A single trial of a bilateral task. reaching to grasp a can with one hand and pull the tab
with the other hand, showing linear wrist velocity (top) and acceleration (middle), and
grip size (lower). Both hands reach the can together, with peak velocity, acceleration
and deceleration occurring earlier and with less amplitude of peak velocity for the
precision hand than for the whole-grasp hand. (Reprinted from Behavioral Brain
Research, Vol. 56, U Castiello, KMB Bennett and GE Stelmark. The bilateral reach to
grasp movement, p. 48, copyright 1993, with permission from Elsevier Science.)
magnitude of trunk and arm motion increasing and elbow motion decreasing
as reaching distance is increased. The lower limbs play an important role in
balancing and supporting the body mass during long reaches (Dean et al.
1999a). Overall, it seems that the trunk plays a general transport role with the
arm playing a corrective role, countering any variability in the movement as
the hand approaches the target (Seidler and Stelmach 2000).
The hand is the principal means of interacting with people and objects in the
environment. Grasp has been categorized as either precision or power (Napier
1956), the former involving pads of digits, the latter the whole hand. Although
this general distinction remains, modern biomechanical research is casting light
on the dependence of grasp configuration on the object and the task to be
Grasping is a complex action involving movement at several joints. The hand
takes on many different configurations in the performance of complex and
skilled manipulations. It is becoming evident, however, that control is simplified
to some extent by certain anatomical configurations and patterns of force pro
duction (Fig. 5.4); for example, many tasks involve interactions between object,
thumb and index finger. The thumb abducts and rotates and the index finger
flexes while force is exerted on the object through the pads of the digits, primar
ily by long flexors of thumb and finger. Each digit is stabilized by a combination
FIGURE 5.4 Task-related grasps showing the different patterns of force production. (Reproduced
from Ibera" et al. 1986, with permission.)
166 Training gUidelines
of joint geometry and intrinsic muscle action. The force produced on the index
finger by the thumb is countered by the first dorsal interosseus muscle and the
force produced on the carpometacarpal joint of the thumb is constrained by
activity of all the thenar muscles. The resulting axial rotary moments of force
are balanced by contraction of other muscles crossing the hand and wrist
(Lemon et al. 1991). The relative contributions of muscles differ and are task
For some tasks, 'locking' (holding cutlery or a tool) and 'supporting' (holding
a loaded plate or tray) grasps involving the 4th and 5th fingers and thumb are
required. These grasps are described in detail by Bendz (1993), but a brief
description taken from Bendz's work is given here.
In the locking grasp (Fig. 5.5a), the implement is locked into the palm of the
hand by the 4th and 5th fingers flexed at all joints and rotated at the metacar
pophalangeal (Mep) joints. The 5th metacarpal is flexed (at the carpometacarpal
joint) by hypothenar muscles, with opponens digiti minimi acting directly on the
metacarpal. Abductor digiti minimi and flexor brevis flex the metacarpal indi
rectly through their common attachment to the proximal phalanges of the little
finger. The major function of abductor digiti minimi is therefore the application
of the flexing and rotational force to the proximal phalanx of the 5th finger. The
force produced by this muscle is augmented by the action of flexor carpi ulnaris
via their common attachment to the pisiform bone. In the supporting grasp (Fig.
5.5b), flexion of the 5th metacarpal enables the little finger to prop up the plate
and keep it horizontal. The work of Bendz is interesting as it examines closely
the contribution of the ulnar side of the hand, in particular the 5th finger, to the
FIGURE 5.5 (a) The locking grasp is used in holding cutlery. The 4th and 5th fingers hold the
implement firmly in the palm. (b) In the supporting grasp, the 4th and 5th fingers are
instrumental in keeping the plate level.
(a) (b)
shaping of the hand. His work also makes clear the importance of specifically
training these two grasps in the contexts of the tasks in which they occur.
Shaping the hand appropriately for any object or task involves the carpal and
metacarpal bones giving the hand a concave shape by the combined action of
intrinsic muscles. Lemon et al. (1991) call this a 'postural set' for the promo
tion of precision movement. A similar concept is that a group of fingers acts as
a single functional unit, a 'virtual finger' (Arbib et al. 1985, Li et al. 1998a,b).
When an object is gripped strongly between thumb and four fingers, forces are
shared between the four fingers in the functional unit, the fingers acting coopera
tively (Li et al. 1998a). Figure 5.4 shows how, in holding a mug, three func
tional units, each producing force in different directions, cooperate to ensure a
stable grasp.
The anatomical structure and the corticomotoneuronal connections to hand
muscles allow a great variety of joint configurations and functional possibilities.
Monosynaptic corticomotor axons diverge intraspinally and make functional
contact with a small number of different target muscles making up the 'muscle
field' of the corticomotoneuronal cell. In playing a musical instrument, the domi
nant pattern may be one of fractionation, or the fingers can move together in a
pattern of co-contraction when force is required as in gripping (Lemon et al.
1991). During strong gripping actions, the contribution of each finger to the
force produced varies according to the finger's position relative to the thumb
(Zatsiorsky et al. 1998). Contributions both to force sharing and to co-activation
of muscles that are not primarily involved in a strong grasp may be mechanical
coupling effects (Kilbreath and Gandevia 1994), with neural factors such as a
widespread neural interaction controlling different muscles (Zatsiorsky et al.
2000). These investigations illustrate how a large number of muscles (a synergy)
can cooperate to produce a required common output (Li et al. 1998b). Since the
object and task are themselves driving the control of the hand, the importance of
patients practising real tasks with real objects is quite evident.
Since shaping the hand for manipulation is partly a function of the anatomy of
the hand, it is critical to preserve the integrity of bony relationships of the
hand (i.e. joint flexibility and muscle length) in the early stages after stroke to
allow recovering muscle activity to be channelled into the ability to hold and
manipulate objects.
While reaching for an object appears to be largely under visual control, once
the hand comes in contact with the object, the major sensory inputs contributing
to movement' control come from cutaneous. tactil .and pressure sensors in the
hand. These -help us identify objects and classify them according to properties
such as texture, shape, weight and potential for slippage, enabling the motor
output (muscle force) to be adjusted accordingly. The two forces produced to
respond to weight of the object and its frictional propertIes are called, respect
ively, the 'vertical lifting force' and the 'slip-triggered grip force' (Johansson
and Westling 1984, 1990, Johansson et al. 1992a,b,c). The extent of sensory
control may, however, depend on the predictability of the task, being used more
intermittently when the object's properties are predictable (Johansson and
Westling 1990).
Another factor of interest is that changes in grip force occur while an object is
being moved in response to the inertial forces produced by movement. Many
patients having difficulty maintaining their grasp on an object while moving
the arm may be demonstrating deficits in adaptive force control, lacking the
ability to adjust force sufficiently rapidly.
There is increasing evidence that recovery is minimal in some individuals, par
ticularly those with an initially severely paretic limb (Nakayama et al. 1994,
Coote and Stokes 2001). Reports of recovery of functional use irrespective of
severity of initial impairment vary from 5% (Gowland 1982) to 52% (Dean
and Mackey 1992). Such studies do not give any indication of the differences
in recovery between those whose initial paresis is severe and those with moder
ate or mild impairment. It is only recently that studies have begun reporting
measures of actual arm use as a real indicator of functional ability (Liepert
et al. 1998, Kunkel et al. 1999). In addition, for results to be meaningful, a study
of recovery needs to provide details of type and intensity of interventions since
the process of rehabilitation itself affects brain reorganization (Liepert et al.
2001) and therefore recovery.
An early report indicated that only 20% of patients with a flaccid upper limb
2 weeks after stroke regained any functional use of the hand (Wade et al. 1983).
There have also been previous reports that absence of a measurable grip by
1 month after stroke indicates poor functional recovery (Sunderland et al. 1989).
According to another recent study (Pennisi et al. 1999), individuals with paralysis
of the hand muscles in the first 48 hours of their stroke, who have an absence of
motor-evoked potentials in response to transcortical magnetic stimulation (TMS),
are likely to have a very poor prognosis. This severely affected group of patients
is the group that typically receives the most intervention, and this may be at the
expense of those with the potential for regaining usable function. Clinical
research reporting the details referred to here is critical in order to be able to
specify interventions appropriately to different groups of patients.
What is not so clear is the potential for recovery of those patients with a mild
to moderate degree of disability. This group may be particularly susceptible to
the methods of intervention used. Several studies have reported positive effects
of interventions involving repetitive exercise and practice of task-oriented and
functionally relevant actions (Sunderland et al. 1992, Butefisch et al. 1995,
Duncan 1997, Kwakkel et al. 1999, Parry et al. 1999b, Nelles et al. 2001), and
evidence comes from a growing number of studies reporting positive effects of
intensive task-oriented exercise of the affected limb during constraint of the
non-paretic limb in individuals with some ability to activate hand muscles (e.g.
Taub et al. 1993, Liepert et al. 2001). However, as Butefisch et al. (1995,)
point out, many interventions still in common use do not emphasize voluntary
activation of the distal arm and hand musculature, but utilize more passive
cutaneous and proprioceptive facilitation techniques and aim for more proxi
mal control of the limb before the hand.
170 Training guidelines
FIGURE 5.6 Functional reorgonization in bilateral motor and sensory systems illustrated by
increases in regional blood flow when the subjects' paretic arms are moved passively.
Areas of activation: (top) before training, (middle) after task-oriented training in the
experimental group, (lower) in the control group. (Reproduced from Nelles et 01.2001,
with permission.)
Preventing the non-paretic limb from reorganizing itself into the 'only available
limb' and forcing activity of the impaired limb may be critical to stimulate the
brain recovery processes that lead to the recovery of effective upper limb func
tion in patients with some active use of the hand. The complexity of hand use
and the imperative to be independent in such functions as eating, drinking and
dressing are factors that need to be taken into account in planning intervention.
As an interesting aside, there is no empirical support for the view espoused by
Brunnstrom (1970) that there is a 'natural' four-stage recovery pattern for the
upper limb in individuals foUowing stroke. Recent studies have shown that the
kinematic patterns of movement used by patients as they reached forward to a
target were due to impaired interjoint coordination and unrelated to the
Brunnstrom stages (Trombly 1992, Levin 1996). Several other studies report
recovery patterns that vary considerably from those observed by Brunnstrom
(Wing et al. 1990, Trombly 1993). Recent research makes it clear that use and
Reaching and manipulation 171
context have an effect on the motor pattern, and the pattern of muscle activ
ity reflects the biomechanical demands of the task and the distribution of mus
cle weakness rather than a stereotyped neurological linkage (Trombly
1993). Nevertheless, these stages form the basis of a measurement tool (the
Brunnstrom-Fugl-Meyer scale) still used in clinical practice and in research
despite its questionable validity (Malouin et al. 1994).
In conclusion, it appears likely that many patients with paralysed hand muscles
early after stroke recover little if any functional limb use. Those with some
hand muscle activity, however, appear to do well if intervention begins early
and emphasizes repetitive exercise, meaningful task-oriented training, inten
sive, forced use of the paretic limb, and bimanual exercise. It is these patients
on whom intensive upper limb training and exercise should be focused.
Impairments such as depressed motor output, decreased rate of neural acti
vation, poor timing and coordination of segmental movements and sensory
deficits impact severely on upper limb functional performance. Muscle weak
ness and loss of manual dexterity may be compounded by the development of
soft tissue changes and shoulder pain, and by the natural tendency for patients
to focus on their non-paretic limb for their daily activities (Carr and Shepherd
2000). Muscle weakness and impaired motor control affect motor performance
according to the distribution of muscles involved. Abnormalities of movement
also reflect the adaptations made as the individual attempts to move.
Research findings
There are relatively few biomechanical and physiological studies of arm func
tion after stroke. Findings from such studies are gradually increasing our
understanding of upper limb function and are helping to explain clinical obser
vations. There is no typical pattern of weakness in the upper limb after stroke
(Colebatch and Gandevia 1989, Duncan et al. 1994). Weakness is no greater
distally than proximally or in extensors more than flexors. There is also no evi
dence that recovery takes place from proximal to distal.
There is relative sparing of some proximal muscles which have bilateral
innervation, such as shoulder adductors (pectoralis major) (Colebatch and
Gandevia 1989). A study of corticospinal influences on two antagonistic muscles
deltoid and pectoralis major - showed that inputs from both hemispheres
appeared to be particularly marked for the adductor muscle, and provided evi
dence of direct corticospinal projections to the deltoid muscle (Colebatch et al.
1990). These results provide some evidence of the possible mechanisms under
lying the clinical observation that some active shoulder adduction is often pre
sent while the converse is true for the deltoid muscle.
Rapidly conducting corticospinal neurons may be preferentially damaged Il1
stroke (Hauptmann and Hummelsheim 1996). Slower recruitment times of
upper limb muscles (flexor and extensor carpi radialis longus, biceps and triceps
brachii) and difficulty sustaining a contraction have been reported (Sahrmann
and Norton 1977, Hammond et a1. 1988). These findings may explain the
slowness of movement and difficulty sustaining a limb position or grasp on an
object commonly observable in patients.
Weakness of glenohumeral joint (GHJ) abductors, flexors and external rota
tors and of supinators has a significant effect on reaching actions, while weak
ness of wrist extensors, finger and thumb flexors and extensors, abductors
and adductors affect the manipulation of objects. Sustaining and controlling
grip force during grasping and lifting objects is a common problem. Force pro
duced during gripping can be slow to build up, difficult to stabilize at the
required level for a specific task and characterized by irregular force changes
(Hermsdorfer and Mai 1996). Deficits in interjoint coordination are evident
during reaching actions, with difficulty making smooth, continuous and accur
ate visually guided arm movements (Trombly 1992, Levin 1996, Cirstea and
Levin 2000). Reaching movements are generally slower than in the able-bodied
(Trombly 1992).
Clinical observation suggests that individual muscles appear weaker for some
actions than for others, i.e. their activity seems to be task dependent (Carr and
Shepherd 1987, Trombly 1993). There is some experimental evidence that this
might be so, and that it might be due to the torques produced by other syner
gic muscles (Beer et a1. 1999). Isometric strength curves in able-bodied subjects
have been shown to change with different limb configurations (Winters and
Kleweno 1993). It is likely, therefore, that biomechanical mechanisms that
stem from segmental interactions may underlie the task- and context-specific
differences in the force produced by a weak muscle post-stroke, particularly if
it crosses more than one joint.
Patients have been shown to demonstrate superior motor performance when
performing a concrete task involving meaningful interaction with an object
compared to an abstract task with no object involved (van der Weel et a1.
1991). In one study, performance was tested as patients either reached forward
to take and drink from a cup filled with water (concrete) task, or performed
the movement without interaction with the object (abstract) task. The move
ment was faster in the concrete task than in the abstract (van Vliet et a1. 1995).
Other studies have shown a significantly greater range of active supination
during a dice game that was performed by rotating a handle (supination) com
pared to turning the handle as a rote exercise without the dice game (Sietsema
et a1. 1993), and superior kinematic performance has been reported during
reaching to scoop coins off the table compared to the same movement without
the coins (Wu et a1. 2000).
Repeated practice of strengthening exercises and functional actions is critical
after stroke as it is for anyone attempting to gain strength and skill in motor
actions. However, the repetitive element is often lacking in physiotherapy prac
tice (Butefisch et a1. 1995) for several reasons, including fear of increasing spas
ticity. The primacy of spasticity as the impairment underlying motor dysfunction
and therefore requiring focus in intervention is still stressed in rehabilitation,
although this emphasis is not supported by clinical evidence. Although reflex
hyperactivity, associated reactions and co-contraction may be present follow
ing stroke, they do not necessarily interfere with function (O'Dwyer et al.
1996, Ada and O'Dwyer 2001). Some currently used therapy approaches are
based on the belief that resisted exercise is contraindicated as it increases spas
ticity. However, neither reflex hyperactivity nor muscle stiffness is increased by
vigorous active exercise. Conversely, these phenomena can respond positively
to vigorous task-specific exercise and training (Miller and Light 1997,
Teixeira-Salmela et al. 1999).
Misconceptions about the clinical signs of spasticity have led to some counter
productive practices. For example, if the presence of muscle activity of finger
flexors early after stroke is assumed to reflect the presence of spasticity rather
than the presence of innervated motor units, active exercise and training may
be avoided in favour of passive inhibitory techniques. In addition, what may
appear on clinical testing to be spasticity (e.g. resistance to passive movement)
may reflect adaptive muscle changes such as increasing stiffness and con
tracture. Clinical tests in use, most of which test resistance to passive move
ment of the upper limb, cannot distinguish between reflex hyperactivity and
adaptive soft tissue changes such as stiffness (e.g. Ashworth scale, Fugl-Meyer
Impairments in sensory perception (in particular tactile and proprioceptive
inputs) and visuospatial impairments can also be major factors that interfere
with reaching and manipulation in some individuals. Control of grip force
depends on afferent inputs from thumb and fingers, and it has been shown
experimentally that grip force regulation is impaired by loss of afferent input
(by digital anaesthesia) to thumb and index finger (Johansson et al. 1992c).
Observational analysis
There are several valid and reliable tests that give objective and functionally
relevant information about upper limb function. A selection is listed at the end
of the chapter. As part of day-to-day motor training, however, therapists must
rely on their own visual observation of motor performance for ongoing analy
sis and as a guide to intervention.
As soon as the person attempts to carry out a voluntary action, adaptive move
ments are evident. The pattern of adaptive movements reflects the pattern of
muscle weakness, the degree of interjoint coordination, the biomechanical pos
sibilities inherent in the segmental linkage, and lack of joint and muscle flexi
bility due to soft tissue length changes and increased muscle stiffness.
Some typical examples of adaptive movement during attempts at arm use are
as follows .
Reaching for an object that is within arm's length: flexion of the upper
body at the hips instead of shoulder flexion (Cirstea and Levin 2000). This
movement has been noted to decrease as shoulder flexion improves.
174 Training guidelines
FIGURE 5.7 (a) Adaptations to weakness or paralysis of shoulder muscles. He uses elevation of
shoulder girdle, lateral f'exion and rotation of spine to swing his arm forward. (b) The
aim is to abduct the thumb to release the glass. Extension of the thumb and wrist
flexion are attempts to substitute for weakness of abductor pollicis brevis.
(a) (b)
Reaching forward: shoulder girdle elevation, lateral flexion of spine,
abduction of shoulder with elbow flexion, internal rotation of shoulder
and pronation of forearm (Fig. S.7a).
Pre-grasp: excessive opening of hand for grasp to compensate for potential
Releasing objects: finger extension with wrist flexed due to contracted long
finger flexors, weak wrist extensors; extending thumb at carpometacarpal
joint (CMC]) and metacarpophalangeal joint instead of abducting at
CMC] (Fig. 5.7b).
Grasping: excessive flexor force in compensation for poor control (see
Fig. 5.13).
In addition, there are three common adaptive sequelae of stroke:
use of the non-paretic limb preferentially when active movement is
possible, and subsequent 'learned non-use'
habitual posturing of the paretic limb, leading to adaptive length-associated
changes to soft tissues including loss of extensibility and increased stiffness
of muscle (Fig. S.8a,b)
Reaching and manipulation 175
FIGURE 5.8 (a) The webspace is reduced compared to normal and the glass forces the thumb into
an unnatural flexed, abducted position at the metacarpal joint. Note the f'exed
wrist. (b) Compare the posture of the hand on the L to the normal posture on the R.
The wrist is flexed with forearm pronated, and 4th and 5th fingers cannot make
contact with the glass. Abnormal wrist-forearm joint relationships appear to have
(a) (b)
joint stiffness and pain, particularly affecting glenohumeral (GH) joint and
Critical components of reach, grasp, manipulation
The complexity of upper limb movement arises from the number of potential
interactions between the intention or goal, the object and the environment. It
is, however, possible to identify components of movement, or sub-tasks, that
are present in tasks that share similar dynamic features.
The simplifications inherent in listing critical components of complex actions are
derived from scientific evidence for task- and activity-related groupings of fin
gers. Despite the apparently limitless variety of tasks performed daily, each task
seems to represent ways of putting together in various combinations a relatively
small number of component parts (Kapandji 1992). These components can be
visualized as configurations, i.e. ways of putting together fingers and thumb to fit
an object and what is to be done with it (Fig. 5.4). This is, of course, an oversim
plification of dexterity, but it is a practical way of dealing with the extensive
needs of the individual with a brain lesion (Carr and Shepherd 1998).
Support for this attitude comes from recent clinical studies which show func
tional improvements after simple wrist and finger exercises (Butefisch et al.
1995). For example, patients who practised repetitive finger and wrist flexion
and extension against various loads twice daily for 15 minutes increased their
grip strength, peak isotonic force of wrist extensions, speed of movement
and functional motor performance. In contrast, a control group of patients
who received neurodevelopmental therapy, with emphasis on decreasing tone
Box 5.1 Training of actions in upper limb function
Forward: flexion at GH joint*
Sideways: abduction at GH joint*
Backward: extension at GH joint* with
- elevation of shoulder girdle
- external rotation at GH joint
- extension of elbow
- supination and pronation of forearm
- extension at wrist
Extension of wrist and fingers, with abduction and conjunct
rotation ('opposition') of the CMC joint of thumb and 5th finger
Flexion of fingers and thumb around object
Extension at wrist
Extension at MCP joint of fingers
Abduction and extension at CMC joint of thumb
Flexion and extension of MCP joints of fingers with wrist in
Palmar abduction and conjunct rotation of CMC joint of thumb
Combined flexion and conjunct rotation at CI"IC joints of 5th
finger and thumb (e.g. cupping)
Independent finger flexion and extension (e.g. tapping)
Key grasp configurations, for example thumb-index; thumb-5th
finger; 4th, 5th finger into palm; thumb + flexed MCP, extended
interphalangeal (IP) joints of fingers (paper-holding grasp)
*These movements at the GH joint are accompanied by movement at shoulder girdle
joints (ratio of glenohumeral to scapulothoracic movement in abduction is 6:1 before and
5:4 after 30 abduction) and, when reaching at arm's length and beyond, movement of the
upper body.
without directly exercising the muscles, did not show improvement in func
tional movement (Butefisch et al. 1995).
Upper limb function basically comprises two groups of actions: reaching/point
ing and grasping/releasing/manipulating. Since extending the distance that can
be reached requires movement of the upper body in sitting and of the whole
body in standing, balancing the body mass is part of upper limb action unless
the individual is fully supported (Ch. 2).
Within these two groups there are key components which can provide a focus
for training and a guide to analysis. Training of actions which comprise these
components has the potential for some generalization to a variety of different
functional tasks commonly performed. Strength training may need to focus on
the muscles that produce these movements (Box 5.1).
In general, the types of action in which we need to be skilled can be compressed
into a short list to aid us in focusing on what is most critical to emphasize in
training. The list below is derived from clinical implications from the research
findings described earlier.
Pick up, grasp and release objects of different shapes, sizes, weights, textures.
Hold and transport objects from one place to another.
Move objects within the hand.
Manipulate objects for specific purposes.
Reach for objects in all directions in sitting and standing.
Use two hands to accomplish specific tasks, e.g.
one hand holding and the other moving (unscrewing lid of jar)
- both hands doing the same movement (rolling out pastry)
- each hand doing two different movements (peeling an apple).
Throwing and catching actions to regain the ability to time action and to
respond quickly to the ball's speed, e.g. throwing ball, bouncing ball, hit
ting ball with bat.
It is now generally accepted that exercise and training need to be specific to
task and context, Le. related specifically to the tasks to be learned. Actions
practised should be concrete rather than abstract, and involve an object
(Wu et al. 2000). Support for task-specific and object-related training comes
from several sources, including studies of constraint-induced training (Coote
and Stokes 2001).
Given the complexity of upper limb functioning and the nature of the neural
lesion, it is a challenge to develop effective methodologies. In individuals
with some volitional motor function early after stroke, it is those
interventions that require active participation and involve repetitive practice
which have been shown to be effective, as Duncan (1997) has pointed out.
The evidence so far supports the following methods of intervention:
Repetitive exercise for wrist extensors, finger flexors and extensors.
(Butefisch et al. 1995)
Forced use (with constraint of non-paretic arm) and intensive exercise
and task training (Taub et al. 1993, Morris et al. 1997, Liepert et al.
Bimanual training (Mudie and Matyas 1996, 2000, Whitall et al. 2000).
Exercise and training appear to be most effective under certain conditions:
Presence of active wrist and finger extension (Wolf's 'Minimum
Motor Criteria') (e.g. Taub et al. 1993)
Sufficient intensity* of meaningful exercise (Sunderland et al. 1992,
Kwakkel et al. 1997, 1999, Parry et al. 1999b, Taub and Uswatte 2000)
"There is some evidence that increasing the intensity of physiotherapy for the upper limb can be
beneficial (e.g. Sunderland et al. 1992, 1994, Kwakkel et al. 1999). However, increasing the
intensity of therapeutic methods which are not particularly effective is unlikely to improve
outcome (Lincoln et al. 1999).
Concrete practice using objects rather than abstract (van Vliet et al.
1995, Wu et al. 2000)
Repetitive exercise (Taub et al. 1993, Butefisch et al. 1995, Duncan
1997, Dean and Shepherd 1997).
There is also evidence of the effectiveness under certain conditions of
electrical stimulation (ES), including functional electrical stimulation (FES)
(Faghri et al. 1994), computer-aided training (Sietsema et al. 1993), and
specific training of sensory awareness (Yekutiel and Guttman 1993; Carey
The guidelines below include: soft tissue stretching to increase the
extensibility of certain key muscles; simple exercises for patients with severe
muscle weakness in the acute stage; and guidelines for organizing training
and practice of everyday tasks involving reaching for and manipulation of
objects, with some specially designed tasks which add interest and
excitement to the practice session. Once patients are discharged, many will
need to continue with exercise and practice. A computer printout with
appropriate exercise suggestions can be supplied. Smits and Smits-Boone
(2000) provide excellent advice on home exercises, including simple
methods of measurement, developed by one of the authors after his stroke.
Occupational therapy is concerned with the practice of functional tasks
involving the upper limbs in real-life environments including at home. Both
physical and occupational therapists need to work together to ensure
consistency in the interventions each provides to the patient.
Major recommendations
1. Increased intensity of meaningful and task-oriented exercise. Intensity of
practice can be facilitated by: group practice, circuit training, work sta
tions; use of assistants as well as physiotherapists to supervise and assist
2. Forced use by constraint of the non-paretic limb with task-specific exercise of
the paretic limb. The unilateral exercises and tasks below can be prac
tised during constraint of the non-paretic limb in patients with some vol
untary activation of wrist and finger extensors and who fail to use their
affected limb outside therapy sessions despite having some functional
use of the limb.
3. Bimanual exercise. Even when a patient has only minimal voluntary activa
tion, bimanual practice (e.g. attempting to open a jar) may have a facilitat
ing effect on muscle activation and coordination of the affected limb
(Mudie and Matyas 2000). One of the reasons for the reported failure of
individuals with a good level of function to use the limb at home (Broeks
et al. 1999) may be difficulty using the two limbs together. Bimanual motor
control differs from unimanual control, particularly in timing parameters,
and needs to be practised for the necessary coordination to be trained.
Repetitive arm training with rhythmic auditory cueing (BATRAC), using a
custom-designed arm training machine, has been reported to improve
motor function, strength and active range of motion immediately after
training and 2 months later (Whitall et al. 2000).
4. Soft tissue stretching. Patients with limited ability to move the limb voluntar
ily to extremes of range need to spend prescribed periods each day with
the limb positioned with muscles at risk of shortening held in a lengthened
position: GH joint internal rotators, adductors; forearm pronators; wrist
and finger flexors; thumb adductor. Patients with the ability to move the
limb voluntarily need to practise exercises which actively stretch these
Soft tissue stretching
Brief passive stretches are carried out immediately before an exercise session
to decrease muscle stiffness and throughout exercises as needed (Box 5.2a).
Active stretching occurs throughout active exercise. The thumb web space
Box S.2a Brief passive soft tissue stretching
Tissue Stretching procedures
Long finger flexors, Brief stretch* with hand on wall, or table top, and
wrist flexors, thumb manually
Forearm pronators Brief manual stretch* with forearm on table top.
Ensure pronator teres muscle is fully lengthened
Adductors and internal Brief manual stretch* in sitting (or supine) with
rotators of GH joint hands behind head (Fig. 5.9b); in sitting with arm
on table; in sitting with arm abducted, externally
rotated, elbow extended
*Hold stretch for approximately 20 s, relax, repeat 4-5 times.
Box S.2b Prolonged soft tissue stretching
Forearm pronators Positioning in mid-supination with forearm on gutter
when in a wheelchair (Fig. 5.23)
Adductor pollicis and Prolonged stretch
using a small plastic wrap-around
web space thumb splint to hold the thumb in palmar
abduction. This allows active grasping and may
prevent contracture in patients with hand muscle
weakness. Practising tasks with a large object in
the hand actively stretches the web space
Adductors and internal Prolonged stretcht in supine with hands behind head
rotators of GH joint (Fig. 5.9b); with arm supported at 90 abduction
on table top, elbow extended, forearm supinated
tHold prolonged stretch for 20-30 min.
180 Training guidelines
(adductor muscle) is stretched during exercises that involve holding objects
of different sizes - the larger the object the greater the stretch.
Patients with severe muscle weakness, particularly around the GH joint,
require periods of more sustained passive stretching and positioning during
the day (Box 5.2b).
Eliciting muscle activity
For patients with severe muscle weakness early after stroke, simple exercises
may elicit muscle activity and increase force-generating capacity. The
exercises may enable the individual to regain the ability to generate muscle
force in a simplified context. Isolated repetitive movements or isometric
contractions of a finger or the wrist may have a facilitatory effect on
neuronal mechanisms (Hauptmann and Hummelsheim 1996).
FIGURE 5.9 ' Stretching and positioning procedures. (a) Brief stretch to pronators before starting
training. (b,c) Positioning to stretch GH adductor and internal rotator muscles.
(a) (b)
Reaching and manipulation 181
FIGURE 5.10 Methods of eliciting muscle activity. (a) An EMG device monitors activity in deltoid
muscle during attempts at passing the cup. (b,c) Demonstration of EMG-assisted
functional electrical stimulation* of paretic wrist and finger extensors (b) showing the
FES electrode over the motor point of wrist extensors with a reference electrode near
the wrist. Between them is the EMG electrode that detects the signal. (c) The aim here
is to push the weight (bean bag) off the table. For triceps brachii, the active FES
electrode is on the R, reference electrode on the L, with the EMG electrode in the
middle. For anterior deltoid the electrodes are similarly arranged but not visible here.
(*Neurotrac 5, Verity Medical Ltd. Photographs by courtesy of Ruth Barker, University
of Queensland, Brisbane.)
(b) (c)
Electromyographic (EMG) biofeedback (Wolf et al. 1994) may be useful in the
context of these exercises to assist the patient to activate a very weak
muscle - the signal from muscles which are too weak to cause movement
provides incentive (Fig. 5.10a). Periods of electrical stimulation (ES), preferably
EMG-triggered, are carried out during the day to selected key muscles: wrist
and long finger extensors (Fig. 5.1 Ob), palmar abductor of thumb, deltoid,
supraspinatus. ES can preserve muscle fibre contractility, and since it can be
done with minimal supervision, it provides a way to increase time spent
182 Training gUidelines
FIGURE 5.11 Simple exercises to elicit muscle activation. (a) In attempts at raising and lowering
the glass, the therapist assists by keeping the forearm in mid-rotation so he can
concentrate on activating his wrist muscles. Eccentric activity may be elicited initially
by attempting to lower the glass. (b) Attempting to move the glass along the table
may initiate activation of wrist extensors and flexors.
(a) (b)
exercising muscles. Mental practice or imagery, in which simple physical
actions are rehearsed 'in the mind', may assist some patients by focusing
attention on the action to be performed (Page et al. 2001).
Simple active exercises
Simple exercises. by exploring the possibilities, may give the patient the idea
of contracting muscles. Wrist extension is critical to grasping, manipulating
and releasing objects.
Sitting with arm on table
Ufting and lowering an object held in palm and (Ingers over end of table
Lifting glass from table by radial deviation at wrist, forearm in mid-rotation
(Fig. 5.11 a), placing it to left and right by wrist flexion and extension
Sliding glass along table top to touch target by extending wrist (Fig. 5.11 b)
Tapping table top with all (Ingers
Discourage wrist flexion unless it is required for the exercise.
Provide targets indicating how far to move, where to place glass.
The natural resting position of the forearm is in pronation. Exercises
are necessary to enable practise of active supination and to retain the
functional length of supinator muscle.
Holding a long ruler, supinate forearm to touch end of ruler to table
Make imprint in putty with knuckles (try for knuckle of index (Inger)
Encourage supination and cupping of hand by pouring seeds etc. with
non-paretic hand into palm
Reaching and manipulation 183
FIGURE S.12 Independent practice of forearm supination and pronation on the Upper Limb
Exerciser. The computer game linked to the manipulandum provides motivational
and quantitative feedback. (Courtesy of Biometrics Ltd, PO Box 340, Ladysmith,
VA 22501, USA.)
Beating a drum (by supinating)
Playing a computer game (Fig. 5.12)
Discourage lifting forearm off table.
Practice of certain tasks is used to train the control of force used in grasp.
The following tasks are also likely to train dynamaesthesia when the patient
pays attention to the shape of the cup.
Holding a polystyrene cup filled with water without deforming it (Fig. 5.13):
transfer it to the other hand; place it on a target
As the cup is moved about, it must retain its shape.
Encourage patient not to spill the water.
184 Training guidelines
FIGURE 5.13 Grasping a polystyrene cup gives feedback about excessive force generation.
Therapist helps by holding forearm in mid-rotation while the patient tries to adjust
muscle forces to avoid deformation of the cup.
Simple exercises to elicit activation of muscles around the shoulder.
In supine. Therapist lifts arm and supports it in flexion, patient attempts
various simple actions:
Reach up to touch target
Take palm of hand to head (eccentric activity of triceps brachii) (Fig. 5.140)
Take hand above head to touch pillow (eccentric for triceps brachii, shoulder
Hand on forehead, move elbow down to pillow and up (Fig. 5. 14b).
If patient experiences some pain when the arm is lifted into position
passively. therapist can apply a minimal amount of traction to avoid nip
ping of stretched soft tissues between joint surfaces.
Shoulder girdle elevation is a critical part of arm movement and inability to
do this action is implicated in poor recovery. The following exercise may
enable the patient to get the idea of the movement.
Reaching and manipulation 185
FIGURE 5.14 Simple arm exercises. (a) Eccentric and concentric activity of elbow extensors as she
takes her hand down to her head and back. (b) Practise of eccentric and concentric
activity of shoulder adductors (pectorals) as she takes her elbow down toward the
pillow and back.
(a) (b)
Sitting with forearm supported. Shoulder shrugging.
Do not allow trunk or head lateral movement as a substitute.
Simple reaching exercises are practised in sitting with arm supported on
table top, for shoulder movement, elbow flexion and extension. These
exercises may stimulate deltoid activity spontaneously in order to reduce
friction of skin on table.
Glass of water in hand, arm on table top.
Slide glass forward in different directions (across the body, out to the side) to
touch targets, keeping forearm in mid-rotation (Fig. 5.15).
Slide glass backward and forwards to touch targets by extending and flexing
In first exercise: elbow should not flex - draw two lines to indicate the
'track' the arm should take.
186 Training guidelines
FIGURE 5.15 He is able to reach out and pass the glass by sliding his arm along the table. A weak
contraction of deltoid, which would normally reduce friction from the table, is elicited
by this movement.
Table top is positioned against the body at shoulder height, shoulder at
90 starting position, varying between flexion and abduction.
Reaching in a more lateral direction is important as it involves some
GHJ external rotation.
When the patient has some muscle activity around the shoulder, reaching
exercises are practised to exercise deltoid (and synergists such as upper
trapezius) in small range movements.
Start with arm on table at 90 shoulder flexion. Reaching and pointing
within controllable range above 90, gradually increasing range, in forward
and sideways directions.
Discourage excessive elevation of shoulder girdle as a substitute for
flexion or abduction.
Include reaching tasks that require mid-pronation or supination of fore
arm, not only pronation.
Discourage overuse of non-paretic limb, such as shoulder elevation.
Objects are placed to 'force' a particular component of movement such
as external rotation of shoulder (Fig. 5.16), mid-pronation of forearm
All practice has a concrete goal, for example, 'Point to touch the target'.
Reaching and balancing practice
Sitting on stool: Reach forward, sideways or backwards to pick up an
object, transport it to another place (e.g. the floor), pick it up again,
Reaching and manipulation 187
FIGURE 5.16 Reaching to pick up a glass. Note glass is placed to 'force' external rotation of GHJ.
FIGURE 5.17 Reaching to open cupboard doors and take out an object can be practised uni- and
188 Training guidelines
reach as far in one direction as possible then put it down. Also practise
bimanually, reaching to pick up large objects that require both hands
(see Fig. 2.10).
Point to different parts of a target, drawing on sheet of paper on wall.
Standing: Reach down to pick up and place objects on stool or floor, vary
distance to be moved according to ability (see Fig. 2.9)
Reaching up to take object from shelf, vary height according to ability
(Fig. 5.17).
Manipulation and dexterity practice
There are many activities that can be used to increase speed and
precision of movement and, as a result, skill. The patient's help should be
enlisted in thinking up tasks to practise that have relevance. Here are some
Tapping tasks
touch each finger tip to thumb in sequence as rapidly as possible
(do a given number of sequences within a given time)
tapping table with single fingers.
Hand-cupping tasks to train opposition of radial and ulnar sides of
the hand
hold seeds in palm and pour into dish
- scooping coins from tabletop into palm of other hand (change
Pick up different objects between thumb and (tnger(s), place them on
various targets
- pick up objects between thumb and 4th, 5th fingers
- pick up small objects from inside a cup with thumb and several
fingers, thumb and forefinger (Fig. 5.18a)
pick up piece of paper from opposite shoulder
- pick up pencil, put it down on table, turn it anticlockwise to point
in opposite direction, then clockwise (Fig. 5.18b); use target lines on
- stack dominoes
- pick up and hold saucer or lid of large jar using 'spider' grip (Kapanji
1992) in which hand spans the whole diameter, thumb extended to
the maximum, fingers stretched wide (Fig. 5.18c).
Pick up larger objects from one side of table and place to other side; vary
weight, distance to be moved
pick up glass of water and drink
pick up jug of water and pour into glass; vary amount of water, size
of jug
':-The items on this list can also be used as a training protocol for patients during constraint of
the non-paretic arm.
Reaching and manipulation 189
FIGURE 5.18 Manipulation and dexterity practice. These tasks train coordinated movements of
fingers, wrist and forearm: (a) picking small objects out of a cup; (b) rotating the pen
through 180; (c) using the 'spider' grip; (d) tracing a circle without touching the lines;
(e) punching in numbers on telephone key pad.
(a) (b) (c)
(d) (e)
- pick up mug of water and drink.
Use a stopwatch to time another member of the group.
More difficult tasks (more complex, or requiring more muscle strength)
type on a computer keyboard, play computer games using
190 Training guidelines
drawing and writing
tracing a circle without touching the lines (Fig. 5.18d)
use telephone keypad (Fig. 5.18e)
pegboard tasks, board games, playing cards
turn door handles, knobs
bounce and catch ball
- turn page of magazine
- lift and move saucepan of water, one handle, two handles
- walk while carrying a glass of water, teacup.
The pads of thumb and fingers are used for grasping, not the lateral
Wrist should be in extension in these tasks.
All opposition movements should take place at CMC joints.
Bimanual practice
As soon as the patient has the ability to control simple movements with the
affected limb, bimanual training should begin (Fig. 5.19). The first two
exercises can be done with minimal muscle activity. If necessary, the hand
can be bandaged to the handlebar.
Arm cycling
Bike riding (arms included)
Push-ups against wall
Pour water from jug to cup/glass and back
Fold a towel
Roll a rolling pin back and forth
Scooping coins off table top into other hand
Remove lids from jars, cans
Pour from one polystyrene cup to another (do not allow deformation
of the cup)
Plunging action with coffee maker against water resistance
Typing on a keyboard - start with two fingers, progress if previously
Remove small objects from pockets
Hold newspaper - turn over pages with paper on table, progress to
holding paper and page turning
Reaching up to cupboard for box, different weights according to ability
Reaching to pick up and place large objects of different shapes and
Walking, walking up and down steps, standing up, holding a loaded tray
Throwing and catching
Manipulating a ball
For the patient to learn to manipulate a particular tool (toothbrush, comb,
tools of trade or recreation) the therapist analyses the patient's
performance of the action to establish what components are preventing
Reaching and manipulation 191
FIGURE 5.19 Bimanual practice: (a) folding a towel; (b) removing the lid from a can requires a
different action compared to unscrewing a lid; (c) pouring from one cup to another;
(d) plunging action; (e) catching keys; (f) arm cycling; (g) rolling a ball between hands.
(a) (b)
(e) (d)
(e) (I) (g)
192 Training gUidelines
Box 5.3 Bimanual training - use of cutlery
Task* Action
Manipulating fork, knife, Forearm in pronation, practise turning small objects
spoon into hand when over in the hand using fingers and thumb
it is picked up Pick up implement and move it into position
Holding fork, knife Pick up and place small objects between thumb and
between ring, little ring or little fingers; supinate and pronate forearm
finger and palm while holding object
Hold paper, putty, etc. between ring and little
fingers and palm - try to pull paper out from grasp
with other hand
Pressing down on fork, Forearm on table with three fingers in a fist, press
knife to hold or down on putty with index finger. Keep finger
cut food extended at IP joints
Drinking from spoon
Hold spoon and carry fluid to mouth. Practise
moving hand while keeping fluid from spilling.
*After initial training, practice these tasks with food.
tThis is a difficult task as the grasp has to be sustained and fluid kept level while the hand
transports the spoon through space.
effective performance. How one might train a patient in the use of cutlery is
described in Box 5.3 as an example.
Strength training
Upper limb muscle weakness is modifiable after stroke in patients with some
muscle activity. It is likely that all patients who regain the ability to generate
muscle force can benefit from strength training, particularly of muscles used
in tasks performed with arms at 90 and above, and those involved in
gripping and holding objects.
Amount and intensity (i.e. amount of resistance and number of repeti
tions) are graded to the individual's ability. As a guide, a maximum num
ber of repetitions up to 10 should be attempted, performed in sets of
Strength training can increase muscle strength without an increase in
Elastic band exercises are progressed by changing to a different coloured
band. Start each exercise with band taut (slack taken up) (Ch. 7).
Some examples
gripping exercises using grip force dynamometer, spring resisted
gripping device or plastic putty
Reaching and manipulation 193
elastic band exercises for GHJ flexors, abductors, external rotators,
elbow flexors and extensors (Fig. 5.20)
exercises with hand weights for wrist extensors and flexors, and
muscles as above
use progressively heavier objects in reaching, lifting and manipulating
FIGURE 5.20 E.lastic bond exercises: (0) exercise for GH] external rotators; (b,c,d) exercises
particularly focused on GH and shoulder girdle muscles.
(a) (b) (c)
194 Training gUidelines
Shoulder pain prevention
As a consequence of weakness, joint immobility, disuse-provoked soft tissue
and joint changes, adhesive arthritis, tendinitis and bursitis may develop,
leading to pain in the shoulder and/or wrist. These inflammatory and painful
conditions may be triggered by injury to the shoulder. Overextensibility of
capsular structures of the GH joint, triggered by a prolonged period with
the flaccid arm dependent, may lead to GHJ subluxation. Decrease in the
pain-free range of movement may occur within the first 2 weeks of stroke
(Bohannon and Andrews 1990), and subluxation within 3 weeks (Chaco and
Wolf 1971). These are common sequelae after stroke and are discussed in
detail elsewhere (Carr and Shepherd 1998, 2000).
Shoulder pain is known to have a negative effect on functional recovery,
impeding rehabilitation (Bohannon et al. 1986, Wanklyn et al. 1996). A major
cause of pain is the development of adhesive capsulitis (Ikai et al. 1998),
which may develop due to immobility coupled with persistent maintenance
of GHJ internal rotation and adduction. I"luscie changes, particularly adaptive
stiffness and shortening of internal rotator-adductor muscles, and weakness
of GHJ external rotator and abductor muscles, are linked to the
development of shoulder pain (Bohannon 1988). Although GHJ subluxation
is assumed to be a cause of pain, there is no evidence that it is (Arsenault
et al. 1991, Ikai et al. 1998, Zorowitz 2001). Pain may, however, be evident
on passive shoulder movement in the presence of subluxation due to
nipping of stretched soft tissues pinched between joint surfaces.
Passive range of motion exercises have been implicated in injury or
activation of previously asymptomatic abnormalities of the paralysed
shoulder (Kumar et al. 1990, Cailliet 1991). Adaptive shortening of muscles
around the shoulder, particularly those linking scapula and humerus
(Fig. 5.21), could interfere with scapulohumeral movement, resulting in small
tears to soft tissues. Impingement of soft tissues may occur if the head of
the humerus is compressed against the scapula when the shoulder is
passively flexed or abducted without GHJ external rotation (Fig. 5.22).
Accidental trauma to the shoulder is implicated in shoulder pain (Wanklyn
et al. 1996) and should be preventable. Patients who need most help in
getting out of bed and standing up from a chair may be particularly
susceptible to injury. Lifting the patient by pulling on the arm appears more
common than one would expect given the likelihood of trauma to a
shoulder unprotected by voluntary muscle activation, with a strong
likelihood of tears to soft tissues due to their poor condition.
Any complaint of pain should be investigated and appropriate intervention
given, as it would be in a non-stroke population. This enables treatment,
such as joint mobilization, to commence. A standardized shoulder pain
prevention programme, however, with particular emphasis on active
exercise performed into external rotation, abduction and flexion-elevation,
and periods of passive stretching, may be the best way to prevent a painful
shoulder developing (Box 5.4).
Reaching and manipulation 195
FIGURE 5.21 Muscles of the shoulder girdle and major forces acting on the shoulder joint.
(Reproduced from Peat 1986, after Dvir and Berme 1978, with permission.)
anterior (lower)
FIGURE 5.22 (a) Scapulohumeral anatomical relationships. (b) External rotation of the humerus
during abduction ensures that the greater tuberosity of the humerus is rotated out of
the way of the acromion process.
(a) (b)
SubluxQtion is most commonly found in patients with extreme muscle
weakness and lack of use of the limb, and must be largely due to the
stretching effects of gravity on inactive soft tissues. It has been thought to be
linked to downward rotation of the scapula positioning the glenoid fossa
more vertically, but recent research suggests this is not the case (Culham
et al. 1995). Decrease in subluxation is related to increases in range of
shoulder abduction and significant motor recovery (Zorowitz 2001).
196 Training gUidelines
Box 5.4 Shoulder pain prevention protocol
Positioning for at least 30 min each day
- in supine: hands behind head (Fig. 5.9b)
- at a table: GHJ in abduction, external rotation (Fig. 5.9c)
Positioning in wheelchair: in mid-GHJ rotation, arm resting on gutter (Fig. 5.23).
Arm must not be positioned in internal rotation for any more than brief
periods during the day
Pain-free active exercises for GHJ external rotators, abductors, flexors
as in guidelines, with emphasis on GHJ between 90 and full elevation
(Fig. 5.18d)
Electrical stimulation: to anterior and posterior deltoid muscle
Avoid activities likely to damage the shoulder, including passive range of
motion exercises, and pulling the patient by the arm
FIGURE 5.23 Arm gutter has been adjusted to ensure the GH joint and forearm are supported in
mid-rotation and the hand is prevented from rolling over at the wrist. Note the
increased tilt shown by the frame on the right which applies pressure against
the wrist.
Reaching and manipulation 197
Slings for shoulder support
A variety of slings and supports are used in the clinic with the hope that they
prevent subluxation and shoulder pain. Although some types of support may
correct an existing subluxation to some extent, there is no evidence that they
prevent subluxation. Most types of sling are contraindicated for two reasons
of significance to the potential recovery of a functional limb: they promote
learned non-use of the affected limb, and they provoke contracture and
stiffness of the GHJ internal rotator and adductor muscles. For these reasons
at least it should be mandatory that a triangular sling or any sling that holds
the GHJ in internal rotation should not be worn. An exception is during
assisted bathing in the early stages, to lessen the possibility of injury.
In the acute stage after stroke, arm support should only be considered when
the limb is paretic, with little or no voluntary activity of muscles around the
shoulder. A cuff-type sling, which allows the arm to hang naturally by the
side, can be worn during sit-to-stand (STS) and walking practice if it makes
the patient feel more comfortable. When sitting, the arm is supported on
the table, or on an arm gutter when in a wheelchair. The arm gutter should
be high enough to prevent the shoulder dropping, and the arm should be
positioned to avoid shortening of the pronator muscles and pronation of the
wrist in relation to the forearm.
Strapping to the shoulder, using non-stretch tape applied over a non-allergic
tape to prevent skin reactions, can provide some support and comfort in
patients who have some active use of the arm (Fig. 5.24).
FIGURE 5.24 Strapping to provide some support for the upper limb. (Courtesy ofJMcConnell.)
Forced use: 'constraint-induced movement therapy'
Taub (1980), after a series of primate studies, described a phenomenon called
'learned non-use' which he proposed may underlie the difficulty some patients
have in using the affected upper limb when discharged home, despite the presence
of a sufficient level of motor control and strength to demonstrate good scores on
clinical tests. One way of forcing limb use in such patients, for which there is evi
dence of effectiveness, is a combination of intensive practice of meaningful motor
tasks with constraint of the non-paretic limb (Taub and Wolf 1997). The ration
ale for this approach is that non-use of a limb is a learning phenomenon involv
ing a conditioned suppression of movement (Taub 1980). Intensive repetitive
practice and constraint may overcome this conditioned response by creating a real
necessity to move. Contrast this with the typical hospital and rehabilitation set
ting, which makes only limited demands on a patient, therefore providing little
incentive to use the affected limb. The patient can manage from day to day by
using the non-affected limb and with the help of staff. Wheelchair mobility,
achieved by single-arm propulsion, may be provided, despite the obvious emphasis
on and over-use of non-paretic limbs, and it is still very common for patients to
hold the affected arm in a sling for a large part of the day.
In several studies of small patient groups (Table 5.1), motor function was shown
to have significantly improved after intensive practice plus constraint of the non
paretic arm in a sling (Taub et al. 1993, Kunkel et al. 1999, Miltner et al. 1999),
with benefits lasting for at least 6 months (Miltner et al. 1999). Most importantly,
many of these studies have shown dramatic improvements in the amount of use of
the affected limb in real-world environments (e.g. Kunkel et al. 1999). Another
study by Taub and Wolf (1997), in which a half glove (fingers free) was used,
showed similar benefits, the patient being taught to use the glove as a cue not to
use the non-paretic extremity. It is important to note that studies with more
equivocal results (e.g. van der Lee et al. 1999) used constraint but without inten
sive task-oriented exercise and training.
Since therapists frequently express concerns about the application of con
straint, it is interesting that a study of patients within 14 days of stroke
reported that the patients tolerated the constraint well, and that no-one with
drew from the programme or lost independence as a result of not being able to
compensate with the non-paretic limb (Dromerick et al. 2000).
The most compelling evidence in support of constraint plus intensive and
meaningful exercise and practice comes from studies of brain reorganization
after stroke. These studies showed an association between increased use of
the affected limb, improved motor performance and brain reorganization.
Liepert and colleagues, using focal transcranial magnetic stimulation, have
reported significant brain changes including enlargement of the hand muscle
output area in the affected cerebral hemisphere, suggesting recruitment of adja
cent brain areas (Liepert et al. 1998, 2000). These changes remained 6 months
TABLES.1 Upper limb function: clinical outcome studies
Reference Subjects Methods Duration Results
Sunderland et al.
Sunderland et al.
Taub et al. 1993
Butefisch et al.
Liepert et al.
132 Ss
<5 weeks post-stroke
Mean age 67.5
(range 32-92) yrs
97 Ss
9 Ss
>1 year post-stroke
Median age 65 yrs
27 Ss
3-19 weeks post-stroke
Age range 35-80 yrs
6 Ss
6 months post-stroke
Mean age 51.9 10.9 yrs
Stratified randomized controlled design
ET: Enhanced therapy (more
intense + motor learning techniques)
CT: Conventional (Bobath) therapy
Separate-groups design
1: CIMT: constraint of unimpaired
arm (splint and sling) + practice of
tasks with impaired limb
2: Attention-comparison group: no
active training (procedures designed to
focus attention on impaired limb +
passive movements)
Multiple baseline design
1: Repetitive resisted exercises to
wrist and fingers
2: TENS for 2 weeks followed by
training as above
Pre-test, post-test design
CIMT: constraint of the unimpaired
arm (hand splint and sling) + intensive
training of impaired limb
ET: 10 weeks
CT: 18 weeks
ET group received>
twice the amount of
Rx per week
14 days
Constraint: 90% of
waking hours
Training: 6 h,
5 days/week
Baseline: 3 weeks
(45 min Bobath therapy)
8 weeks
15 min/twice daily
Constraint: no details
Training: 2 weeks
6 h/day, 5 days/week
At 1 month: ET had significantly greater
improvement on Extended Motricity Index
(EMI) compared with CT p < 0.01.
At 6 months: no significant difference between
the groups; however, ET-mild subgroup had
significantly better arm function on EMI, Nine
Hole Peg Test p < 0.01
Frequency of arm pain in ET > CT
1-year follow-up: improvement not retained
1: Significant decrease in time on Emory Motor
Function Test, in time taken to perform a series
of tasks (Arm Motor Activity Test)
2: No improvement on any test
2-year follow-up: gains were retained
Baseline: no change in either group
1: Significant increase in grip strength p < 0.006,
peak isometric wrist extension force p < 0.05,
peak acceleration of isotonic wrist extension
p < 0.05
2: No improvement
Motor function improved significantly on
Amount of Use Test (AOU), Motor Activity Log
(MAL). Significant increase in amplitude of
motor-evoked potentials in affected
hemisphere, and in size of cortical area capable
of eliciting activity of abductor pollicis brevis
p < 0.05
_L-'-{C_o_n_t_in_u_e_d__ ) _
Reference Subjects Methods Duration Results
Lincoln et al.
Parry et al. 1999
Whitall et al.
Blanton and
Wolf 1999
Miltner et al.
Kunkel et al.
282 Ss
recruited <5 weeks
Median age 73 yrs
As above
16 Ss
>12 months post-stroke
Age range 44-89 yrs
1 S
4 months post-stroke
15 Ss
>6 months post-stroke
Mean age 54
(range 33-73) yrs
5 Ss
> 3 years post-stroke
Mean age 53
(range 47-66) yrs
Single blind randomized controlled
RPT: Routine Bobath therapy
QPT: Additional Rx with qualified
APT: Additional Rx with assistant
As above
Single group design
Bilateral repetitive arm training with
rhythmic cueing (BATRAC)
Single case design
CIMT: constraint of unimpaired
hand (mitten) + practice of tasks
Baseline, pre-test, post-test design
CIMT: constraint of unimpaired arm
(hand splint and sling) + training
Pre-test, post-test design
CIMT: constraint of unimpaired arm
(splint and sling) + training
5 weeks
RPT: 30-45 min,
5 days/week
QPT and APT: 10 h
additional Rx
As above
6 weeks
20 min, 3 times/week
14 days
Constraint: all waking
Training: 6 h/day,
5 days/week
12 days
Constraint: 90% of
waking hours
Training: 7 h/day for
8 days
14 days
Constraint: 90% of
waking hours
Training: 6 h/day for
10 days
99 Ss completed study
No significant effect on arm function
No between-group differences on Action
Research Arm Test (ARAT), Rivermead Motor
Arm Score (RMAS) at 3 and 6 months
When outcome data in less severely impaired
group were analysed, APT showed significant
increase in ARAT and RMAS compared with
Significant improvements in Fugl-Meyer Upper
Extremity Test, Wolf Motor Function Test
p < 0.05
2 months follow up: improvements retained
Improvements on Wolf Motor Function Test
(WMFT) at post-test. Amount of Use Test
(AOU) on Motor Activity Log (MAL) at
3-month follow-up: further improvements on
No change between baseline and pre-test.
WMFT time score significantly decreased.
Increased use of impaired arm (AOU on MAL)
6-month follow-up: functional gains retained
WMFT time score significantly decreased.
Actual Amount of Use (AAUT) increased 98%
3-month follow-up: improvement retained
_ ~ (C_o_n_t_in_u_e_d__ )
Liepert et al.
Leipert et al.
Nelles et al. 2001
13 Ss
>6 months post-stroke
Age range 42-68 yrs
9 Ss
4-8 weeks post-stroke
Mean age 59.7
(range 50-72) yrs
10 Ss
<22.8 days
E: 68.0 3.6 yrs
C1: 63.0 3.0 yrs
Baseline, pre-test, post-test design
CIMT: constraint of unimpaired arm
(splint and sling) + training
Baseline t(Jtl: 1 week 'conventional'
therapy, no details
Forced-use tl-t2: constraint of
unimpaired hand (forearm splint) +
'conventional' therapy
Randomized controlled design
E: task-oriented training
C 1: passive movements, stretches,
soft tissue mobilization
C2: 5 age-matched healthy volunteers
12 days
Constraint: 90%
waking hours
Training: 6 h/day for
8 days
1 week
Constraint: duration
fixed individually
a priori
3 weeks
45 min per session
Baseline: no change in Motor Activity Log
Pre-test to post-test: significant improvement
on MAL (p < 0.0001). Transcranial magnetic
stimulation (TMS) paralleled functional results
6-month follow-up: functional improvement
Cortical area size became almost identical in
both hemispheres
to and t1: no change in Nine Hole Peg Test. TMS:
no change in motor output area
tl-t2: Nine Hole peg test completed in
significantly less time (p < 0.02), grip strength
increased (p < 0.00). TMS: motor output area in
affected hemisphere significantly larger than in
unaffected hemisphere
E: Serial PET imaging revealed functional brain
reorganization (relatively more activation
bilaterally in the inferior parietal cortex (IPC)
and in the contralateral sensorimotor cortex)
C1: Weak activation of ipsilaterallPC
C2: No change in age-matched healthy
C, control: CIMT, constraint-induced movement therapy: E, experimental; PET, positron emission tomography; Rx, treatment; Ss, subjects; TENS, transcutaneous electrical
nerve stimulation.
Patients for whom this intervention should be provided include those who pass
certain 'minimum motor criteria' (Taub and Wolf 1997) but who do not use
the affected limb in daily tasks despite ability to move the limb voluntarily.
This is the group of patients reported to benefit so far. It is not yet known
whether patients without the minimum motor criteria would benefit. However,
one patient in Taub and Wolf's investigations, who did not meet the minimum
motor criteria, did not improve in activities of daily living.
Constraint of the non-paretic limb in almost all the studies reported so far
involved a resting hand splint and sling worn for a large part of each day (90% of
the waking day except for certain agreed upon functions) for 14 days. There is
some evidence that wearing a mitt or glove may also provide effective constraint
but it must be accompanied by staff actively discouraging use of the limb. The
advantage of a mitt or glove is that it allows the limb to be used should safety be
compromised. It should be noted that in these investigations parients attended for
an intensive 6 h of task-oriented exercise and training. A shorter period may also
be effective and future studies should address this question.
The Taub group ensure the patients' understanding and compliance with the
constraint by:
1. teaching them to put on and take off the splint and sling independently
2. making a behavioural contract with each patient, working from the daily
routine (whether in or outpatient) to decide which activities are to be
carried out with constraint and which with two hands (see Morris et al.
1997 for details)
3. requesting a diary be filled in to enable monitoring of patient's activities
outside therapy time.
In the protocol in Box 5.5 we have outlined a methodology following Taub
and Wolf's research methodology. This protocol could be followed for 2 weeks
initially and continued for longer if necessary.
Computerized training
Computer games controlled by a modifiable lever or manipulandum are likely to
be increasingly used in the training of various aspects of upper limb movement.
Efficacy has been reported in a training study of a 13-year-old boy with Erb's
palsy (Krichevets et al. 1995). Use of a game in this manner focuses attention on
outcome of movement rather than the movement itself, and the motivating
effects of being an active participant in an interesting task may be powerful facili
tators. A computer is also used in strength training for particular muscle groups.
A hand-strengthening device is connected to a computer so that particular mag
nitudes of force production are required to operate the game (Pashley 1989).
There have been several encouraging reports showing efficacy (King 1993).
Figure 5.25 shows the clinically significant changes in active supination range of
movement after a period of computer-assisted training in one individual (shown
in Fig. 5.12) following stroke.
Reaching and manipulation 203
Box 5.5 Recommended protocol in constraint-induced movement therapy
Criteria for inclusion
Behavioural contract
Programme of exercise
Wolfs minimum motor criteria: ability to
voluntarily extend wrist at least 20
, IP and MCP
joints of fingers at least 10
Willingness of a patient to attend initially for a
training programme 5 days/week for 2 weeks
Hand splint with wrist of non-paretic arm in some
extension + sling, or mitt/glove + reminders not
to use the non-paretic limb
Wear for a prescribed number of hours on each
weekday for 2 weeks
Seek agreement with patient about activities to be
performed without constraint (e.g. bathing, parts
of dressing, activities where balance may be an
issue) and with constraint (e.g. eating, grooming,
toileting, some domestic tasks)
Prescribed number of hours of task-oriented
training and practice for the affected limb + home
Before and after 2 weeks
FIGURE 5.25 The results for one individual (see Fig. 5.12) after 6 weeks' training of supination with
a computerized system. (Unpublished data courtesy of Dr S Kilbreath.)
0, 60
c 40
Training with computer' commenced
Baseline I
(receiving standard physiotheraphy)t
April 15 April 17 April 19 April 21 April30 May 5 June 10
'Computer-assisted training: 2 sessions of 14 minutes per day
Electrical stimulation
There is increasing interest in the potential of electrical stimulation (ES) to
reduce secondary adaptive changes in muscle, to initiate muscle activity, and to
prevent stretching of rotator cuff muscles and the GH] capsule. ES enables
repetitive muscle exercise in those individuals who have little active movement
in the early stages (Chae et al. 1998, Chae and Yu 2000). This repetitive exer
cise provides a means of retaining the contractility of muscle fibres.
According to two comprehensive reviews (Chae and Yu 2000, Price and
Pandyan 2001), the evidence of increases in pain-free range of passive external
rotation of GH] and reductions in severity of subluxation supports the use
of neuromuscular ES for short-term goals. Functional electrical stimulation
(FES) - the use of multiple contraction sites to evoke a particular pattern of
muscle activity - has shown positive results following stroke. For example,
stimulation of supraspinatus and posterior deltoid muscles prevented shoulder
pain and subluxation, and improved external rotation of GH] when stimulation
was administered for several hours a day (Faghri et al. 1994, Chantrain et al.
1999). Intramuscular electrodes implanted into supraspinatus and posterior del
toid have been reported to have similar effects (Yu et al. 2001). Improvements
in active wrist and finger extension have been reported and decreases in resist
ance to passive movement and co-contraction (Chae and Yu 1999, Powell et al.
1999, Cauraugh et al. 2000). However, evidence of generalization into improved
functional activity remains limited. Although arm function may improve in the
short term there may be no long-term maintenance effect (Sonde et al. 2000). It
is also noteworthy that beneficial effects of ES are maintained after cessation
of treatment only when patients have regained some functional use of the limb
(Yu et al. 2001).
Combining afferent biofeedback from an EMG signal with ES can be effective
(Francisco et al. 1998, Cauraugh et al. 2000). For example, improvements in
manual dexterity and on the Motor Assessment Scale have been reported in
individuals with some voluntary wrist extension (Cauraugh et al. 2000).
However, another study compared EMG-triggered ES and repetitive hand
exercises in individuals with some active limb use. It found no improvements
in grip strength or maximum isometric and isotonic wrist extension force after
the 2 weeks of ES but improvements in all tests after the 2-week hand exercise
programme (Hummelsheim et al. 1997). The authors pointed out that ES
required no cognitive engagement which is a critical factor in motor learning.
This may partially explain the fact that the latter study found repetitive volun
tary exercise of hand muscles superior to EMG-triggered ES. Few studies have
addressed a combination of task-oriented training and ES to enhance motor
learning but the results are promising (Mackenzie-Knapp 1999).
Further clinical research will clarify the patients who may benefit, what these
benefits may be and optimal dosages. In the future, neuromuscular ES using
active repetitive movement training mediated by cyclic and EMG-triggered ES
may prove to be effective as an aid to task-specific motor training and repeti
tive exercise in promoting motor learning in certain patients.
__ Tests for upper limb function*
Test Reference Testing method
Functional tests
Motor Assessment Scale:
upper limb items
Nine-hole Peg Test
Grip force
Spiral test
Arm Motor Mobility
Test (AMAT)
Action Research Arm
Test (ARA)
Measures of real-life
arm use
Motor Activity Log:
Amount of Use
Scale (AOU)
Actual Amount of Use
Test (AAUT)
Biomechanical tests
Shoulder range of
motion tests
Tests of isometric strength
Tests of sensation
Carr et al. 1985, Poole
and Whitney 1988,
Malouin et al. 1994
Mathiowetz et al. 1985
Mathiowetz 1990,
Bohannon et al. 1993,
Hermsdorfer and
Mai 1996
Verkerk et al. 1990
Kopp et al. 1997
van der Lee et al. 2001
Taub et al. 1993
Taub et al. 1993
See Trombly 1993
Mngoma et al. 1990
Lincoln et al. 1998
Carey 1995
The test consists of 8 separate motor items, each measured
on a 7-point scale. Three items measure upper limb function:
upper arm function, hand movements, advanced hand
activities. Administration is strictly standardized. For patients
who achieve top scores, additional tests of dexterity need to
be performed such as the NHPT below
A measure of dexterity. Time taken to complete the test is
measured as the patient grasps 9 pegs and places them in
holes on a board
Grip-force dynamometer
A measure of coordination. Patient draws a line as quickly as
possible between two spirals, separated by a distance of
1 cm, without touching either spiral. Particularly useful for a
patient with cerebellar ataxia
Measures ability to perform 13 activities of daily life
composed of 1-3 component parts. The time taken to
perform each task is measured with a stopwatch; the actions
are videotaped and rated on 6-point scales
The ability to grasp, move, release objects of different size,
weight and shape is tested by measuring three sub-tests
(grasp, grip, pinch) on a 4-point (0-3) scale. Points 2 and 3
are qualitative in the original scale but subjectivity can be
overcome by setting time limits for each item
Provides information about actual use of the limb in life
situations. Patient reports at semi-structured interview
whether and how well, on a 6-point (0-5) scale, 14 daily
activities were performed during a specified period
Measures actual use of the limb on 21 items using a 3-point
rating scale. Patients are videotaped
Kinematic analysis of reaching
Isokinetic dynamometry: LIDO Active System
, a valid
and reliable measure of resistance to passive external
rotation of the GH joint
Goniometer plus hand-held dynamometer (to standardize
Hand-held dynamometry; grip force
dynamometry; pinch
Nottingham Sensory Assessment
Tactile Discrimination Test
Proprioceptive Discrimination Test
*Conditions of testing must be standardized.
tLoredan Biomedical Inc., 3650 Industrial Boulevarde. West Sacramento. CA, 95691, USA.
tDigital Pinch/Grip Analyser, MIE Medical Research (see Sunderland et al. 1989).
There is a large number of tests for upper limb function (see Wade 1992, Carr
and Shepherd 1998). Table 5.2 provides a selection of commonly used func
tional measures which are valid and reliable, and some biomechanical testing
Impairments and adaptations
Strength training and physical conditioning
Adaptive motor patterns
Sensory impairments
Somatosensory im
Visual impairments
ive features
logical, mechanical and
nal changes in soft tissues
r impairments
In clinical practice, the relevance of spasticity remains controversial. Clinical
research findings are often difficult to interpret due to:
the use of different measurements
confusion arising from the persistent use of the word 'tone', a concept
with no clear meaning (Niam et al. 1999)
Following an acute upper motoneuron (UMN) lesion, it is generally recognized
that the major impairments interfering with functional motor performance are
paralysis and weakness (decreased muscle force) and loss of dexterity (dis
ordered coordination) (Landau 1980). In addition, adaptations occur in response
both to the neural lesion and to subsequent disuse (i.e. to joint immobility and
lack of muscle contractility). Common soft tissue adaptations to disuse which
have the potential to impact severely and negatively on the potential for regain
ing functional motor performance include increased muscle stiffness and muscle
shortening or contracture.
Although spasticity - defined as velocity-dependent stretch reflex hyperactivity
or hyperreflexia (Lance 1980) - can be demonstrated in some patients following
stroke from 4-6 weeks onwards (Chapman and Weisendanger 1982, Thilmann
et al. 1991a, Thilmann and Fellows 1991), its significance for functional per
formance and the causative mechanisms of the hyperreflexia remain equivocal.
There appears to be no evidence that a relationship exists between spasticity
(reflex hyperactivity) and degree of disability in voluntary movement (Fellows
et al. 1994), and the contribution of spasticity to motor disability following
stroke may therefore be smaller than previously thought (Pierrot-Deseilligny
1990). The contribution of increased resistance to passive muscle lengthening
(called hypertonia) may, however, be considerable.
Impairments and
the synonymous use of the terms hypertonus (resistance to passive
movement) and spasticity (reflex hyperactivity).
It is important that future clinical studies define the terms used so it is clear
what is actually being tested.
What is not controversial, however, is that, despite a long-standing focus on
spasticity in both research and rehabilitation, recent findings indicate that the
major impairments in terms of their effects on motor performance and functional
activity are muscle weakness and loss of motor control, and that adaptive
changes to muscles and other soft tissues are major factors impacting negatively
on recovery. Somatosensory impairments and disorders of perceptual-eognitive
function can also impact on functional activity in some individuals. In the follow
ing discussion, we present some recent research-based information on impair
ments and adaptations which has relevance to clinical practice and which clarifies
the direction of physiotherapy intervention at the present state of our knowledge.
It has been typical to classify the features of the 'upper motor neuron
syndrome' as positive and negative (Jackson 1958, Burke 1988) (Fig. 6.1). Since
it is becoming clear that certain adaptations follow on from the lesion-induced
features, Figure 6.1 includes a third set of features which can be identified from
experimental studies as adaptive (Carr and Shepherd 1998). This classification
has some explanatory value for clinical practice in providing guidance as to
mechanisms underlying clinical signs. However, in reality the human system is
so naturally flexible and adaptive that it is difficult to draw a line between the
primary impairments and some of the secondary adaptations. Historically, the
FIGURE 6.1 Classification of the positive and negative features of the upper motoneuron lesion
(UMNL) and their relationship with the adaptive features.
Positive features
UMNL Adaptive features
Negative features
Muscle + connective tissue changes
(altered mechanical and functional properties)
Hypertonus (resistance to passive movement)
Altered motor patterns
Loss of dexterity
negative and positive features have been considered relatively independent phe
nomena, related to site and amount of tissue damage and spontaneous recovery
processes (Landau 1980). However, it is now considered that many of these fea
tures are adaptations to lesion-induced impairments compounded by adapta
tions to disuse subsequent to the lesion. Further research will no doubt enable a
more cogent set of descriptors.
Muscle weakness
Following stroke, muscle weakness arises from two sources: primarily from the
lesion itself, as a result of a decrease in descending inputs converging on the
final motoneuron population, and hence a reduction in the number of motor
units available for recruitment. Since skeletal muscle adapts to the level of use
imposed upon it (Lieber 1988), secondary sources of weakness arise as a conse
quence of lack of muscle activity and immobility (Cruz-Martinez 1984, Farmer
et a1. 1993). Contrary to previous opinion, weakness in agonist muscles is not
due to spasticity (reflex hyperactivity) in an antagonist muscle group (Sahrmann
and Norton 1977, Gowland et a1. 1992, Newham and Hsiao 2001) but is a
direct result of reduction in descending motor commands, compounded by dis
use and adaptive muscle changes.
Normally the production of muscle force is dependent upon the number and
type of motor units recruited and the characteristics of both the motor unit dis
charge and the size of the muscle itself. Muscle force is augmented by increasing
the number of active motor units and the firing rates and frequencies of those
units. The structural features of muscle (e.g. cross-sectional area) also contribute
to the potential force which can be produced. Strength is therefore a neuromus
cular phenomenon. As a result, a neural lesion in which motor commands to
the muscles are deficient results in a decrease in strength, even paralysis.
Intact descending inputs are necessary for shaping complex movements by
graded activation of coordinating muscles and for bringing motoneurons to the
high frequency discharge necessary for sustained contraction strength (Landau
1988); i.e. for muscles to produce the force required to carry out a goal
directed action, a necessary number of muscle fibres must contract simultane
ously and the various muscles involved in the action must coordinate their
force production specific to the task and context. The interruption of descend
ing pathways following stroke results in a decrease in the number of motor
units activated, decreased firing rate of motor units and impaired motor unit
synchronization (Rosenfalck and Andreassen 1980, Gemperline et a1. 1995).
These factors cause disorganization of motor output at the segmental level and
underlie the motor control problems exhibited by patients even when they are
able to generate some force.
The distribution of muscle weakness varies among individuals, probably reflect
ing such factors as lesion location and size. Colebatch and Gandevia (1989)
examined arm muscle weakness in two individuals, one with a greater distal than
proximal weakness, the other with a similar degree of involvement of all muscles
tested (Fig. 6.2). Another study (Adams et a1. 1990) examined the distribution
212 Appendices
FIGURE 6.2 Distribution of muscle weakness in two subjects with hemiparesis. The subject shown
in the top bar chart shows greater distal than proximal weakness; the subject in the
lower chart shows similar involvement in all muscles tested. Ad, adduction; Ab,
abduction; FI, flexion; Ex, extension; FDI, first dorsal interosseus. (Reproduced by
permission of Oxford University Press from Colebatch ]G and Gandevia SC 1989.
The distribution of muscular weakness in upper motor neuron lesions affecting the
arm. Brain, 112, 749-763.)
iii 80
~ 40
Shoulder Elbow Wrist Fingers Thumb Grip FDI
Ad Ab FI Ex FI Ex FI Ex FI Ex
~ 40
Shoulder Elbow Wrist Fingers Thumb Grip FDI
Ad Ab FI Ex FI Ex FI Ex FI Ex
of weakness in the lower limb in 20 individuals and found the distribution did
not conform to a common pattern in different subjects, nor was any single
muscle group always the most severely affected. With the group overall there
was a tendency for greater weakness of muscles crossing the ankle than those
crossing the hip and knee, with no significant difference between degree of
weakness of flexors compared to extensors. A second study of 31 individuals
supported these findings (Andrews and Bohannon 2000). Several studies have
reported weakness of muscles of the limb ipsilateral to the lesion (Adams et aJ.
1990, Desrosiers et al. 1996, Harris and Pol key 2001).
The above findings demonstrate that, contrary to the common clinical view,
muscle weakness does not follow a particular typical pattern following stroke.
There was no evidence of consistently greater distal weakness than proximal
or of flexion versus extension weakness, and no difference between severity of
weakness in upper versus lower limb (Duncan et a1. 1994, Andrews and
Bohannon 2000). There is relative sparing of muscles of the trunk (abdominals
and erector spinae) after stroke, in part due to their bilateral innervation
(Kuypers 1973), and several clinical studies have reported minimal reductions in
strength of trunk muscles (Dickstein et al. 1999). Weakness of trunk muscles
does not, therefore, appear to be the major problem following stroke as sug
gested in some therapy literature (e.g. Davies 1990).
After an UMN lesion, weakness is demonstrated by deficits in generating force
and sustaining force output (Bourbonnais and Vanden Noven 1989, Davies et al.
1996). However, the reduction in peak muscle force (torque) is compounded by
a decrease in the rate of force development (Canning et al. 1999). In experimental
work on muscle activation, impaired initiation and cessation of muscle activity as
well as inability to sustain a contraction have been reported (Hammond et al.
1988). Control problems may vary among different muscles. For example, in a
study of nine subjects following stroke, delayed recruitment and inability to sus
tain a contraction were found in extensor carpi radialis while flexor carpi radialis
had delayed recruitment but could maintain a contraction (Hammond et al. 1988).
Slowness of movement and in building up tension during the initiation of move
ment can have a major effect on functional performance, are more evident in fast
than in slow movements (Hammond et al. 1988, Farmer et al. 1993), and appear
to be associated with impaired motor unit synchronization. Canning and col
leagues (1999) cite a patient who could generate a peak elbow flexor torque of
9 Nm, sufficient to lift her forearm and carry a small object. However, since it
took her 7 s to reach peak torque, the muscle's apparent strength was of limited
functional use. Several clinical studies have reported slower-than-normal per
formance of actions such as walking and sit-to-stand (Giuliani 1990, Ada and
Westwood 1992).
An interesting phenomenon, which is clearly observable in the clinic, is the pres
ence of differential degrees of muscle weakness according to joint position and
task. For example, a person may be able to generate sufficient muscle force to
bear weight in standing without knee collapse when the knee is flexed a few
degrees but not be able to hold the knee extended to 0 (Winter 1985, Carr and
Shepherd 1987). One inference is that the knee extensor muscles are able to
generate sufficient force when the muscles are at one length but not at another.
The above example may be explained by findings from research into muscle
function. In normal muscles it has been shown that strength is affected by the
length at which the muscle is contracting. The muscle's length affects actin and
myosin overlap and the length of the moment arm (Zajac 1989, Winters and
Kleweno 1993). Strength-torque measures show that most torque is normally
generated at mid-length and least at the shortest length (Herzog et al. 1991a,b).
A differential effect of length on strength, with selective weakness of elbow flex
ors and extensors at shorter lengths, has been found experimentally following
stroke (Ada et al. 2000).
Tests of muscle strength (Ch. 7) include tests of isometric strength using a
dynamometer (van der Ploeg et al. 1991, Andrews et al. 1996, Andrews and
Bohannon 2000) or the modified Medical Research Council muscle tests, called
the Motricity Index (Collin and Wade 1990). The Lateral Step-up test (Worrell
et al. 1993) provides an indication of functional lower limb strength. Use of
such tests requires close observance of standardization procedures.
Loss of dexterity
Loss of dexterity appears to result from deficits in the sustained and rapid trans
fer of sensorimotor information between cerebral cortex and spinal cord
(Darien-Smith et al. 1996). According to Bernstein, dexterity is the ability to
carry out any motor task precisely, quickly, rationally and deftly (cited in Latash
and Latash 1994). He suggested that dexterity is not present in the motor act
itself but rather in its interaction with the changing environment. Although the
term 'dexterity' is commonly associated with the hand, Bernstein's definition
enables its use to encompass all skilled motor activity.
Loss of dexterity appears to involve the loss of coordination of muscle activity
to meet task and environmental requirements through an impaired ability to
fine tune coordination between muscles (Kautz and Brown 1998). Although
impaired dexterity is typically considered in its association with weakness and
slowness to contract muscle, there is some evidence that they may be independ
ent phenomena (Canning et al. 2000).
Lack of consistency in timing and modulating muscle activation may be a
cause of the decreased coupling between muscle activation and target move
ment reported when stroke patients attempted a flexion-extension elbow
movement to track a target with a low friction manipulandum (Canning et al.
2000). Impaired fine tuning of muscle coordination has also been noted during
ergometer peddling (Kautz and Brown 1998). Since rate of transmission from
cortex to spinal cord affects dexterity (Darien-Smith et al. 1996), individuals
following stroke may have difficulty adapting to the reduced number of corti
cospinal inputs, having to rely also on slower and less direct corticobulbar
channels for relaying information (Canning et al. 2000).
At the neurophysiological level, the relationship between muscle strength (gener
ation of appropriate force) and dexterity (coordination of muscle activation) is
unclear. However, even if they are independent phenomena, in a functional
sense it is likely that they are linked, being both task and context specific and
necessary for skill. Control of synergic linkages (i.e. timing, generating and
coordinating the necessary forces for a particular context) is probably the neces
sary factor that underlies dexterity. There is physiological evidence that, for the
forearm, cortical motoneurons can project to the motoneuron pools of more
than one muscle.
A relationship has been reported between increased muscle strength in lower
limbs and improved standing balance (Bohannon 1988) and walking speed
(Sharp and Brouwer 1997, Teixeira-Salmela et al. 1999, 2001), and between
hand muscle strength and functional hand movements (Butefisch et al. 1995).
Such results may reflect improved coordination after exercise, and demonstrate
that improved force production and coordination of muscle activity according
to task and contextual requirements are both required for effective motor
Box 6.1 provides a summary of muscle weakness and loss of dexterity.
Box 6.1 Summary of muscle weakness
Muscle strength is normally dependent on the number and type of motoneu
rons (MN) recruited and characteristics of MN discharge. Structural features
of muscles contribute
Muscle weakness after UMN lesion is due primarily to a decrease in descend
ing inputs converging on spinal MN and secondary disuse
Dexterity is the ability to perform a motor task in a coordinated manner
Decreased number of MN activated, decreased firing rate and impaired motor
unit synchronization lead to disorganization of motor output at segmental
level and underlie motor control problems (including loss of dexterity)
- weakness is not due to spasticity of an antagonist muscle
- there is no typical pattern of weakness, no consistent difference between
proximal and distal muscles, upper and lower limb, flexors and extensors
- trunk muscles are relatively spared
A widely accepted definition of spasticity is that of a motor disorder character
ized by a velocity-dependent increase in tonic stretch reflexes with exaggerated
tendon jerks, resulting from hyperexcitability of the stretch reflex (Lance 1980).
However, the term 'spasticity' is typically used in the clinic in a more generic
sense to cover a disparate collection of clinical signs including hyperreflexia
or reflex hyperactivity, abnormal movement patterns, co-contraction and hyper
tonia. The term 'hypertonia' is usually reserved for a perceived resistance to
lengthening of muscles by passive movement.
This imprecise use can create a problem in reading clinical studies, since it may
not be clear whether it is stretch reflex hyperactivity that is being measured or
physiological and mechanical changes to muscles such as increased stiffness. Since
the measurement and intervention methods used in clinical practice reflect the
clinician's understanding of the mechanisms underlying a clinical sign, it is import
ant to attempt some clarification. The ensuing discussion examines the possible
mechanisms underlying the clinical phenomena traditionally considered to be
indicative of spasticity, in an attempt to more usefully guide treatment planning.
Reflex hyperactivity
It is not clear to what extent hyperreflexia is present in patients after stroke, or
what its contribution to functional disability might be. Although there are several
studies that report hyperreflexia some time after the lesion (Thilmann et al.
1991a), others have reported stroke subjects with no significant increase in tonic
stretch reflexes during velocity-controlled stretch (O'Dwyer et al. 1996, Ada et al.
1998), and no electrical activity in muscles of limbs in which there was resistance
to passive movement.
Increased reflex activity may not be evident clinically for 4-6 weeks after stroke
(Chapman and Wiesendanger 1982), and some studies have demonstrated
that it can increase progressively over time. These findings suggest that reflex
hyperactivity may be an adaptive response (Burke 1988, Brown 1994). Muscle
extensibility and tonic stretch reflex hyperactivity appear to be closely linked
(Gracies et al. 1997). The tonic stretch reflex is modified not only by the velocity
of stretch but also by the length of the muscle at which stretch occurs (Burke et al.
1971). The activation of tonic stretch reflexes may occur earlier in range (at a
shorter length) than usual, even if hyperreflexia is not present (O'Dwyer et al.
1996, Ada et al. 1998). It is theoretically possible therefore that the gradual
development of stretch reflex hypersensitivity could be influenced by increasing
muscle stiffness and contracture in patients who are relatively immobile; i.e.
hyperreflexia may occur as an adaptive response to non-functional, contracted
and stiff muscles (Gracies et al. 1997) which are persistently activated at short
lengths over a long period. A report that reflex hyperactivity could only be
elicited toward the end of range in a muscle with contracture suggests the pres
ence of contracture may be a potentiator of the stretch reflex (O'Dwyer et al.
1996). It is also suggested that increased stiffness in connective tissue that
attaches to muscle spindles may increase sensitivity (Vandervoort et al. 1992).
An interesting clinical observation is that muscles most at risk of shortening
(e.g. shoulder adductors, internal rotators, ankle plantarflexors) are also those
traditionally thought to develop spasticity.
Resistance to passive movement
The classic view is that increased resistance to passive movement (also called
hypertonia) is indicative of spasticity, and arises as a result of exaggerated stretch
reflex activity (Lance 1980). Theoretically, the mechanisms underlying this
clinical sign have been considered to include reflex hyperactivity and peripheral
factors such as changes within the musculoskeletal system, including increased
muscle stiffness and decreased muscle length. It has been assumed that velocity
dependent stretch reflex hyperactivity can contribute to resistance to passive
movement because the two (resistance and hyperreflexia) occur together in some
patients (Thilmann et al. 1991a).
It is becoming increasingly evident, however, that factors other than reflex
hyperactivity (e.g. adaptive increases in muscle stiffness, changes in muscle
length, structural reorganization of connective tissue) may be major contribu
tors to resistance (Dietz et al. 1981, Hufschmidt and Mauritz 1985, Ibrahim
et al. 1993, Sinkjaer and Magnussen 1994, Malouin et al. 1997). The issue of
adaptive changes is discussed in more detail later in this chapter.
The assumption that increased resistance to passive movement is an indicator
solely of spasticity can be confusing, particularly when reading the results of
clinical research. For example, tests commonly used in the clinic which are said
to measure spasticity, such as the Ashworth scale (Ashworth 1964), directly
measure resistance to passive movement (Pandyan et al. 1999) and any reflex
component is not distinguishable. This is demonstrated by the report of a signi
ficant relationship between Ashworth scale scores and muscle stiffness (Becher
et al. 1998). The pendulum test, also used to measure spasticity, similarly does
not distinguish between reflex hyperactivity and muscle stiffness (Fowler et al.
1997). Since such tests do not differentiate between the two phenomena, they
cannot identify the cause of the resistance (Neilsen and Sinkjaer 1996).
Lack of resistance to passive movement ('hypotonia') may be largely the result
of weakness preventing voluntary muscle activation (van der Meche and van
Gijn 1986). The historical explanation that the presence or absence of resistance
is due to increased or decreased tone does not add to our understanding since
the concept of tone has itself no clear meaning (Niam et al. 1999).
Excessive and redundant muscle activity
Clinical signs such as abnormal movement patterns, muscle co-contraction and
associated movements have been assumed to reflect spasticity. However, recent
research has made it clear that there could be other explanations. Several studies
of elbow flexors have shown no association between reflex hyperactivity and
unnecessary muscle activity (Bourbonnais and Vanden Noven 1989, Canning
et al. 2000, Ada and O'Dwyer 2001). There is some evidence that excessive mus
cle activity may be related to abnormal electromyography-torque relationships
(Dietz et al. 1981, Bourbonnais and Vanden Noven 1989) and reflect adaptive
changes in synergic relationships (Carr and Shepherd 1998, 2000). Abnormal
muscle activation may also result from altered descending central nervous system
(CNS) influences, such as an increase in facilitatory input to alpha motoneurons,
or from secondary peripheral changes such as increased sensitivity of alpha
motoneurons to synaptic input due to changes in neurotransmitter concentration
(Singer et al. 2001).
Excessive muscle activity is frequently apparent in muscles that act unopposed due
to weakness of their usual synergists. For example, the way in which the patient
moves when attempting a task with the upper limb illustrates the effects of
unopposed muscle activity on the dynamics of the segmental linkage. A com
monly observed adaptive movement pattern seen in individuals with muscle weak
ness as they attempt to reach forward for an object illustrates this point (Fig. 6.3).
In the presence of marked deltoid weakness, the patient may flex the shoulder by
contracting biceps brachii. If so, the elbow may also flex because the biarticular
biceps imposes a force on the forearm which is normally countered by activity of
biarticular triceps. If triceps is paretic, it cannot act effectively as an antagonist. It
is pertinent to note that electromyography (EMG) of two-joint muscles in
able-bodied individuals is maximal when a muscle is acting as an agonist over
both joints. In the clinical example above, biceps - since it is unopposed by its
antagonist - is acting concentrically as an agonist over both shoulder and elbow.
It is particularly common to see repetitive and excessive activity of this muscle.
Co-contraaion ofmuscles
This is not itself an abnormal phenomenon in motor control. Muscle co
contraction and stiffening of a limb can illustrate lack of skill and is found in
able-bodied individuals attempting a novel task, or moving at fast speeds, and
in children. Excessive muscle activation following stroke may therefore be a
manifestation of lack of skill in reorganizing task-specific muscle activation
patterns in the presence of inadequate motor unit recruitment and weakness.
A decrease in motor unit firing rate (Gemperline et al. 1995) results in decreased
tension, and additional motor units have to be recruited to enable greater force
218 Appendices
Biceps brachii flexes the elbow as well as the shoulder when triceps brachii
;s paralysed.
development. The result for the patient can be an increased sense of effort at low
levels of activation (Tang and Rymer 1981). Stiffening the lower limb in stance
may illustrate difficulty in controlling muscle contraction when bearing weight
through the limb. A patient may stiffen the limb, generating excessive force in
several muscles in order to prevent limb collapse. Functionally inappropriate
co-contraction can also be due to prolonged activation of a muscle, for example,
carried over from agonist phase to antagonist phase in a reciprocal movement
such as elbow flexion and extension. Nevertheless, several studies have found no
co-contraction (Davies et al. 1996) in stroke subjects performing simple actions.
Associated movements
Also called synkinesis or overflow, these movements are frequently seen in abJe
bodied individuals under conditions of stress. They are recognized as redundant
movements or increases in muscle activity which are also evident in children
younger than 5 years (Abercrombie et al. 1964). In adults they are associated
with lack of skill when performing complex tasks or when generating maximum
force levels (Carey et al. 1983). They have been defined as unintentional move
ments which may accompany volitional movement, resulting from irradiation
or overflow of neuronal excitation in the cortex or spinal cord during voluntary
Associated movements are also seen in patients after stroke during stressful activi
ties, most noticeably in the upper limb. The term 'associated reactions' may be
used when the redundant movements occur in the presence of pathology (Walshe
1923) and are defined in some physiotherapy literature as pathological movements
indicative of spasticity and occurring in the presence of spasticity with increased
effort (Bobath 1990, Stephenson et al. 1998). As a result, when such movements
are observed during strong muscle contraction in another limb, they may be con
sidered by the therapist to be contraindications to exercising against resistance for
fear of eliciting associated reactions and thereby increasing spasticity.
A recent report found associated movements in the affected biceps brachii in
7 out of the 24 patients tested (Ada and O'Dwyer 2001). The amount of muscle
activity was related more to the the amount of torque produced during 50%
maximum voluntary contraction (MYC) of the contralateral quadriceps femoris
than to contraction of the contralateral biceps brachii. There was no relationship
between associated movements and spasticity (tonic stretch reflex hyperactivity).
It is notable that other studies reporting a relationship between associated move
ments and spasticity (Dickstein et al. 1996, Dvir et al. 1996) have measured resist
ance to passive movement (Ashworth scale) and not tonic stretch reflex activity.
Persistent limb posturing
Persistent posturing such as the sustained flexion of the elbow observed in
some individuals after stroke with no signs of soft tissue shortening or hyper
active stretch reflexes (O'Dwyer et al. 1996), may be due to neural and/or
mechanical factors as yet not understood. These factors may include length
dependent facilitation of elbow flexors related to increased stiffness or shorten
ing of muscle fibres associated with an adapted resting length of biceps caused
by prolonged positioning of the elbow in flexion. There has been little investi
gation of muscle resting length and it is not well understood, although it seems
likely that different muscles normally have different resting lengths.
Muscle inactivity with persistent joint (limb) posturing can be associated with
changes in spindle sensitivity, muscle activation being triggered at a shorter
length than normal (Gioux and Petit 1993). It is possible that the signal for limb
position may be provided by the resting discharge of muscle spindles (Gregory
et al. 1988). If the sense of the limb's position is affected, this could account for
the altered resting position ('abnormal posture') noticeable, particularly at the
elbow, in some patients following UMN lesion. It is possible that in some indi
viduals chronic hyperactivity of muscles at a shortened length may have a com
pounding effect on stretch reflex hypersensitivity (Gracies et al. 1997).
Since there is no evidence of a relationship between persistent limb posturing
and spasticity, such posturing is more likely to reflect the history of the muscles,
such as gravitational effects, limb immobility, soft tissue disuse changes and
functional adaptations to spindles and to muscle resting length. The distorted
dynamic posturing seen in some individuals - termed 'dystonia' (Fig. 6.4)
may, however, result from a different mechanism.
The role of stretch reflexes in functional actions is therefore unclear and if
hyperactive reflexes are present after stroke they do not necessarily constitute a
major impairment (O'Dwyer et al. 1996, Becher et al. 1998). Reflex hyper
activity does not appear to be a major contributor to tension development in
muscles during active movement (Dietz and Berger 1983, Berger et al. 1984),
220 Appendices
FIGURE 6.4 Dystonic movement evident during a reach to pick up a glass. Note the pronated
forearm, flexed wrist and extended metacarpophalangeal joints.
although it may contribute to muscle stiffness when the joint is moved passively
(Vattanasilp et a1. 2000). It is also not clear to what extent spasticity is adap
tive nor to what extent muscle (and muscle spindle) changes impact on the
neural system and contribute to stretch reflex hyperactivity. However, clinical
signs such as increased stiffness, hyperreflexia and abnormal movement patterns
may reflect, at least in part, adaptation of the system to the neural lesion and
to subsequent inactivity and disuse.
It has been shown recently that sustained passive stretching for 10-15 min can
decrease motoneuron excitability (Hummelsheim and Mauritz 1993), with the
effect lasting longer when stretch is prolonged to 48 h with plaster casting.
However, an isolated reduction in stretch reflex hyperactivity is not necessarily
followed by improved motor performance or by the ability to learn a new motor
skill. Exercise and motor training, however, can have positive effects on func
tional performance whether or not spasticity is present (Wolf et a1. 1994), and
active exercise can itself cause a reduction in muscle stiffness, co-contraction and
associated reactions as well as an improvement in functional performance
(Butefisch et al. 1995). Muscle strength can be increased by resisted repetitive
exercise without an increase in resistance to passive movement or associated
reactions, but with improvements in functional performance (Miller and Light
1997, Sharp and Brouwer 1997, Teixeira-Salmela et al. 1999). Botulinum toxin,
administered by injection directly into stiff muscles, can reduce resistance to pas
sive movement where this is severe. However, as with any technique that has a
local effect, it is likely that the benefits achieved in the short term can only be
maintained by ongoing stretching, strength training and task practice.
Whether or not a focus on measuring reflex hyperactivity is valid after stroke
is being questioned (Vattanasilp and Ada 1999, Pomeroy et al. 2000). Tests in
clinical use that are purported to measure spasticity (Ashworth scale, pendulum
I Box 6.2 Summary of spasticity
Spasticity is characterized by a velocity-dependent increase in tonic stretch
reflex activity
Reports of the presence of reflex hyperactivity in individuals after stroke are
Spasticity may be an adaptive response. It is evident 4-6 weeks post-stroke,
increasing over time. It is modified by muscle length
Spasticity appears unrelated to functional disability in stroke and its contribu
tion to motor disability may be minimal
Major causes of resistance to passive movement are adaptive muscle stiffness
and contracture. The role of reflex hyperactivity (spasticity) is unclear - it may
contribute in some individuals
Abnormal patterns of movement, co-contraction and redundant muscle activity
may be, at least in part, adaptations to muscle weakness, changes in motoneu
ron firing patterns and soft tissue contracture
Note: Abnormal movement patterns, excessive and redundant muscle activity
(associated movements, co-contraction) are not necessarily indicative of spas
ticity but reflect adaptations to muscle weakness. Strength training is not
contraindicated by the presence of these phenomena
test, Fugl-Meyer scale) do not measure reflex hyperactivity but resistance to pas
sive movement, and no relationship has been found in studies comparing the
Ashworth scale with instrumented tonic stretch reflex evaluation in calf muscles
(Vattanasilp and Ada 1999). Measures of velocity-dependent reflex activity
require equipment generally available only in a laboratory.
Resistance to passive movement, however, is an indication of such mechanical
factors as stiffness and contracture. It can be tested by the modified Ashworth
scale, which is reported to be reliable if a standardized protocol is used
(Bohannon and Smith 1987). The modified Tardieu scale (Tardieu et al. 1954,
Gracies et al. 2000) also evaluates resistance to passive movement. It specifies
the velocity of movement, and grades the muscle reaction by measuring the
angle at which the catch response is detected. It also quantifies clonus.
Range of motion and muscle stiffness can be measured using a dynamometer and
goniometer with a standardized protocol. This has been described for the ankle
(Moseley and Adams 1991), using a specially designed instrument which allows
standardization of the magnitude and angle of applied force (see Fig. 7.4).
Box 6.2 provides a summary of spasticity.
Physiological, mechanical and functional changes in soft tissues
Adaptation can occur at all levels of the neuromusculoskeletal system from
muscle fibre to motor cortex (McComas 1994). The potential of adult skeletal
muscle fibres to change their biomechanical properties in response to func
tional demands is remarkable (Dietz et al. 1986). Lack of muscle activity and
of joint movement results in adaptive anatomical, mechanical and functional
changes to the neural and musculoskeletal systems. These include loss of func
tional motor units (McComas 1994), changes to muscle fibre type, physio
logical changes in muscle fibres and changes in muscle metabolism, and
increased muscle stiffness. Changes in joints include proliferation of fatty tissue
within joint space, cartilage atrophy, weakening of ligament insertion sites and
osteoporosis (Akeson et al. 1987). Malalignment of joints can also occur when
muscles are weak and a limb is not in active use. Malalignment of the wrist
seems to be common. Figure 5.8b illustrates the rotation of the hand into a
pronated position at the wrist, which can develop into a change in bony align
ment of the hand in relation to the forearm.
Paretic muscles have reduced oxidative capacity and decreased overall endurance
(Potempa et al. 1996). Compared to the less affected leg, the paretic limb after
stroke shows reduced blood flow, greater lactate production and a greater util
ization of glycogen plus a diminished oxidative capacity (Landin et al. 1977).
Changes to muscle fibre type include severe atrophy of type II muscle fibres and
predominance of type I fibres (also atrophied) during later stages (Edstrom and
Grimby 1986). There are suggestions that changes in muscle discharge pattern
may cause the muscle fibre transformations (Dietz et al. 1986).
Weakness and fatiguability can arise secondarily from disuse, especially if the
person spends much of the day sitting down. Secondary changes are particu
larly evident in antigravity postural muscles since their previous roles are no
longer practised. Quadriceps atrophy has been reported as early as 3 days after
immobilization in non-stroke individuals (Lindboe and Platou 1984), with a
30% reduction in cross-sectional area within 1 month (Halkjaer-Kristensen and
Ingemann-Hansen 1985). Disuse accentuates fatigue by further reducing the abil
ity of motor centres to recruit motoneurons maximally (McComas et al. 1995).
Impaired motor drive can be a result both of the lesion and disuse, loss of motor
drive also being reported in individuals without brain damage (McComas
1994). Adaptive changes to the cardiovascular system arise from lack of phys
ical activity, and fatigue can be a sign of a low level of aerobic fitness as well as
of low muscle endurance.
Increased muscle stiffness and contracture
Imposed immobility after stroke is associated with changes to muscle occurring
within a few days. Increased passive stiffness *, due to adaptive mechanical and
morphological changes in muscle fibres, appears to be a major cause of disabil
ity in many individuals following stroke (Sinkjaer and Magnussen 1994), and
is a major contributor to the increased resistance to passive stretch found on
clinical testing. Patients themselves complain of 'stiffness' interfering with their
attempts at active movement. Clinical and laboratory tests have demonstrated
*Stiffness can be defined as the mechanical response to a tensile load on a non-contracting
muscle (Harlaar et al. 2000). It is the ratio of the tension developed in the muscle when it is
stretched to the length of the muscle (Herbert 1988).
the functional interference imposed by increased stiffness (Sinkjaer and
Magnussen 1994). Increased stiffness in paretic plantarflexors has been
reported by 2 months post-stroke (Malouin et al. 1997), although it is likely to
occur much earlier as has been shown in animal experiments (Herbert and
Balnave 1993). Severe stiffness has been reported more than 12 months follow
ing stroke (Sinkjaer and Magnussen 1994, Thilmann et al. 1991b). Postural
muscles with a large proportion of slow twitch fibres (e.g. soleus) appear par
ticularly vulnerable. Stiffness and contracture of soleus can have significant
functional effects, interfering with walking, standing up and balancing in
There is increasing evidence that stiffness results primarily from adaptive
changes in mechanical fibre properties of muscle and tendon (Hufschmidt and
Mauritz 1985, Thilmann et al. 1991b, Carey and Burghardt 1993) as well as a
build-up of connective tissue (Williams et al. 1988). These changes occur due to
lack of contractile activity. The length of immobilized muscle affects the
changes that occur over time. Muscles subjected to prolonged positioning at a
short length, and which are rarely exposed to active or passive stretch, undergo
changes in crossbridge connections, lose sarcomeres and become shorter
(develop contracrures) and stiffer. Arthrogenic structures appear to play an
increasing role in limiting joint motion, particularly if immobility extends
beyond 2 weeks. Changes in connective tissue contribute both to stiffness and
contracture (Williams et al. 1988), since water loss and collagen deposition also
occur in response to disuse (Carey and Burghardt 1993). Increased stiffness is
also found in the absence of contracture (Malouin et al. 1997).
The phenomenon of thixotropy is a possible contributor to stiffness and resist
ance to passive movement (Hagbarth et al. 1985). A substance is called thixo
tropic if it can be be changed from a stiff gel to a fluid solution after stirring;
i,e. stiffness of a substance such as muscle is reduced by motion but is
re-established some time after motion ceases (Walsh 1992). Results from a recent
study of 30 individuals with stroke suggest that thixotropy may not be a major
contributor to long-term stiffness but confirmed that contracture was a signifi
cant contributor (Vattanasilp et al. 2000).
Whether or not there is a neural contribution to passive muscle stiffness is
unknown. Sinkjaer and Magnussen (1994) describe stiffness in a contracting
muscle as the sum of the non-reflex features, i.e. the properties of fibres con
tracting prior to the stretch (intrinsic properties) and the response from passive
tissues, together with the reflex response from stretch-mediated reflex contrac
tion. They tested a group of nine individuals post-stroke by stretch to the calf
muscles imposed on top of a maintained contraction and found that the reflex
and intrinsic contributions to stiffness were within the range of the healthy
subjects tested, while the passive stiffness was significantly greater.
Adaptive motor patterns
Adaptive motor patterns are evident after stroke (Carr and Shepherd 1987,
O'Dwyer et al. 1996), and are usually (except when there is dystonia) functional
Box 6.3 Summary of adaptive features
Decreased muscle activity and joint movement leads to adaptive anatomical,
mechanical and functional changes in the neuromuscular system
Changes to muscle resulting from weakness and disuse include altered muscle
fibre type and length. atrophy and altered metabolism. Functional sequelae are
increased stiffness and weakness. decreased endurance and fitness
Increased muscle stiffness is a major contributor to resistance to passive move
ment and a major cause of disability
Adaptive motor patterns reflect muscle weakness. imbalance. stiffness and
adaptations seen during attempts at goal-directed movement. These motor
patterns appear to reflect an individual's best attempt at an action given the
state of the neural and musculoskeletal systems and the dynamic possibilities
inherent in the musculoskeletal linkage (Figs 5.7, 6.3) (Carr and Shepherd
1998). Adaptive movements during attempts at functional actions illustrate the
muscle imbalance caused by paralysis or extreme weakness in some muscles
and not in others, i.e. by the unopposed activity of stronger muscles. Soft tissue
contracture also has an impact on performance, preventing the required range
of movement at a particular joint.
Box 6.3 provides a summary of adaptive features.
Concluding this section, at the present time the research literature supports the
current shift in emphasis in physiotherapy toward addressing weakness, loss of
dexterity (lack of coordination) and the adaptive changes occurring to muscle
and connective tissue. The former require exercise and training to increase
strength (force generation) and promote motor (re)learning; the latter require
passive stretching, active stretching in exercise and training, with, if necessary,
electrical stimulation, splinting or botulinum toxin injection. There is evidence
that such an emphasis can have positive effects on muscle contractility,
strength and functional motor performance. Definitive studies of the effects of
stretching, exercise and training on the development of adaptive muscle
changes such as stiffness, contracture and stretch reflex sensitivity have yet to
be carried out. There seems to be no evidence that the prevailing emphasis in
physiotherapy on decreasing spasticity has any effect on functional perfor
mance, nor that it is necessary to inhibit hyperactivity before exercise and
training (Wolf et al. 1994).
Partial or complete loss of discrete sensation (tactile or proprioceptive) and
perceptual-eognitive disorders (visuospatial impairments, inattention) impact
upon motor behaviour to varying degrees. Difficulties with discriminating and
interpreting sensory inputs have the potential to affect in particular the recovery
of hand function and balance. For example, poor feedback from the glass
while taking a drink, the knife slipping in the hand when eating, or unsure
whether or not the knee will collapse while going up a step, contribute to lack
of confidence and may result in a decrease in activity and, in the case of the
hand, an adaptation to one-handed living.
For the purposes of this book we include only a brief resume of this topic. The
reader will find more information in an earlier publication (Carr and Shepherd
1998) and in the references cited below.
Somatosensory impairments
Loss of discrete sensation represents a failure of sensory impulses to reach the
relevant areas of the brain from the various sense organs of skin, joints,
muscles, ears, eyes and mouth (see Carey 1995 for review).
Light touch, pin prick and temperature are usually recognizable, although the
qualitative element may be only crudely appreciated. A lesion of the brain stem
may result in hemianaesthesia and/or hemianalgesia on the contralateral or ipsi
lateral side, depending on the lesion site. Spontaneous pain (sometimes acute)
down the opposite side of the body may result from sub-thalamic lesions.
Although impairment of joint position and movement sense may for some
patients result in a failure to recognize that movement is occurring, others may
be able to recognize limb movement but not the position of the limb or direc
tion of movement. Tactile impairments include difficulty with localization and
discrimination of tactile inputs causing astereognosis (inability to identify a
common object by its physical properties with eyes closed), poor two-point
discrimination (inability to recognize two points when simultaneously applied
with eyes closed) and inability to recognize bilateral simultaneous stimuli.
Sensation functions in both regulatory and adaptive modes (Gordon 2000),
guiding movements during their execution and correcting movements in order
to improve the next attempt. Recent experimental work has focused on the
relationship between motor and sensory function during the performance of
real-life tasks and there is now a substantial body of knowledge of relevance to
physiotherapists, enabling the development of strategies for testing and for spe
cific training of sensory discrimination and motor performance. Of particular
interest is work on the relationship between sensation and functional motor
output in the hand. For example, Johansson and Westling (1984) have shown
in healthy subjects that grip force varies according to the friction properties of
the object being handled, in that the more tendency there is for the object to
slip the more force is generated by the finger flexor muscles.
There have been several studies of individuals with neural lesions that illustrate
the impact sensory impairments can have on functional performance, particu
larly on hand use. A study of 21 people following stroke found a high correl
ation between perception of joint position and the ability to produce coordinated
limb movements (Lee and Soderberg 1981). Other studies of individuals with
a sensory impairment associated with a cortical lesion have described motor
impairments including slowed movements, difficulty in performing coordinated
finger movements, difficulty in sustaining a constant level of force and poor
spontaneous use of a limb, particularly of the hand (Jeannerod et al. 1984,
Dannenbaum and Dykes 1988).
Training specific somatosensory perceptions can be effective (Carey et al. 1993,
Yekutiel and Guttman 1993). Whether or not this improvement in sensory per
ception generalizes into more effective functional motor performance is currently
unclear. It is likely that the task-specific training practices described in this text
also impact on sensory discrimination since the nervous system is selective in
utilizing those sensory inputs that are most relevant to the action. Practice of
meaningful tasks and specific exercises may give the damaged system the oppor
tunity to regain the ability to recognize, select and use those sensory inputs that
are relevant to the action being practised. For example, practice of standing up
and sitting down provides the opportunity to attend to and utilize inputs from
tactile receptors in the soles of the feet.
Challenging and meaningful problems posed to the hand as a sense organ may
be addressed by the patient attending to and concentrating on the form of
sensory inputs and their relationship to the task being attempted (Yekutiel and
Guttman 1993). Training can involve cueing the individual into the sensory
information required for the task. Encouraging the patient to pay attention to
the relevant sensory cues (e.g. visual, tactile, muscle force), setting up an environ
ment appropriate to the task, and providing verbal and visual feedback as
reinforcers may not only affect motor performance but may also improve sensory
awareness. Paying attention appears critical. Animal studies have shown that
focusing attention increases the responsiveness of cells in sensorimotor cortex
(Hyvarinen et al. 1980).
The impairment that the system needs to overcome involves the interpretation
of signals from the sense receptors that are reduced or distorted by the brain
lesion. This is why providing arbitrary sensory stimulation (icing, brushing,
sensory bombardment) to which a patient need not attend is unlikely to result in
improved use of sensory inputs during motor functions. Such stimulations may
be seen by the system as mere noise (Yekutiel and Guttman 1993).
Where there is a specific sensory impairment, such as inability to recognize
bilateral simultaneous tactile stimuli, or astereognosis, specific training can be
provided to assist the person to regain the maximum sensitivity possible. Such
training may not need to be modality specific (Wynn-Parry and Salter 1975),
which may result in poor generalizability, but rather may require the solving of
meaningful sensory-identification problems.
Several studies illustrate the potential for sensory retraining involving cognitive
manipulation of sensory information. One controlled trial of the training of sens
ory function involving the hand demonstrated that somatosensory impairments
could be improved even years after stroke (Yekutiel and Guttman 1993). The
authors supported the incorporation of systematic sensory retraining as well as
motor training into rehabilitation. Sensory training involved 45-minute sessions
three times a week for 6 weeks based on the following principles.
The nature and extent of sensory loss was explored with the patient.
Emphasis was on sensory tasks the patient could do and each session
started and ended with these.
Tasks which interested the patient were chosen.
Use was made of vision and the non-paretic hand to teach tactics of
Frequent rests and changes of task were used to maximize concentration.
Tasks to practise included: identification of number of touches or lines, and of
numbers and letters drawn on the arm and hand; 'find your thumb' when
blindfolded; discrimination of shape, weight and texture of objects or materials
placed in the hand. Subjects were tested on location of touch, sense of elbow
position, two-point discrimination and stereognosis. Only patients in the treat
ment group made large gains on the sensory tests. Further research is needed
to examine whether or not improved sensory awareness is associated with
improved functional performance, particularly whether patients were using
their affected hand more spontaneously.
Another study (Carey et al. 1993) examined a programme of task-specific train
ing carried out by a small number of subjects 5-26 weeks following stroke,
using attentive exploration and quantitative feedback. The results showed a
marked improvement in tests of tactile and proprioceptive discrimination which
was maintained in most subjects for several weeks.
It is not known if specific sensory training can affect reorganization processes
within the brain itself. Animal experiments have shown that somatosensory
cortical representations of skin surfaces are remodelled by use throughout life
and, following cortical damage, cortical maps have been shown to reorganize
following training (Jenkins et al. 1990).
Visual impairments
Vision is a major source of information about the environment and our place
in it. Reaching out to pick up an object involves complex interactions between
head, eye and hand movements. The action is guided by visual information
about the location of both object and arm, and about the relationship between
wrist, finger and hand movements just prior to and while grasping the object
(Mountcastle et al. 1975, Jeannerod 1988). Visually selective neurons are driven
by particular object properties such as shape and orientation (Taira et al. 1990).
Vision is also crucial for identifying and negotiating critical features of the
environment as we walk about, features such as obstacles to be stepped over,
and traffic lights which require us to time our walking across the street.
Visual impairments can have a negative impact on the patient's ability to engage
actively both in rehabilitation and in daily life. Impairments can include pre
existing retinal dysfunction, and problems with visuocognitive manipulations
of inputs to the retina manifested by such clinical signs as visual field loss or
visuospatial agnosia.
Perceptual-cognitive impairments
Perceptual-cognitive function enables the processing and interpretation of sens
ory information. Perceptual-cognitive impairments reflect a breakdown in the
interactive links between perceptual and cognitive processes. These impair
ments include inattention, disorders of right-left discrimination, or body image
and lack of awareness of one (usually the left) side of space. There are a num
ber of manifestations of the last two. In 1953, Critchley classified nine types of
impairment, pointing out that they are not clearly circumscribed but fuse into
one another. These impairments include unilateral spatial neglect, lack of con
cern over the existence of hemiplegia, denial of hemiplegia with delusions
(anosognosia) and loss of awareness of one body half (autotopagnosia). The
agnosias are taken to reflect not only a failure of sensory perception but also of
cognitive awareness.
Unilateral spatial agnosia or 'neglect'
Individuals who do not identify, respond to, or orient toward meaningful stim
uli which are contralateral to the lesioned hemisphere, but whose deficit cannot
be attributed to sensory or motor impairment, are said to demonstrate neglect
of the body and extrapersonal space contralateral to the lesion site (Ladavas
et al. 1994). This agnosia is most commonly reported following a lesion of the
right hemisphere; however, it is evident that some degree of visuospatial mal
function may also exist with left hemisphere lesions (Wilson et al. 1987).
Spatial disorders can involve impairments of perception, attention, memory
and executive function (Calvanio et al. 1993). Neglect may be accompanied by
discrete sensory impairment (tactile, proprioceptive and stereognostic), muscle
weakness or paralysis, impaired coordination and visual field deficits. Patients
frequently exhibit reduced general arousal and inability to sustain attention.
Reported incidence of unilateral spatial agnosia varies from 29-85% of
patients (Wilson et al. 1987), probably depending on tasks tested and on the
definition of neglect. Agnosia may be transitory, with individuals making a
spontaneous recovery within a few weeks (Wilson et al. 1987). However, it may
persist in some individuals despite lack of obvious behavioural indications.
Severe and persistent neglect is considered to have serious effects on rehabilita
tion and recovery (Kinsella and Ford 1980, Calvanio et al. 1993), which may
reflect the co-occurrence of neglect with large severe strokes. There have been
few systematic investigations and no clear consensus on the evaluation of
Neglect can be manifested as a 'directional hypokinesia', in which the impair
ment appears to be a defective organization of arm movement toward the left
side of space (Bottini et al. 1992). Spatial and temporal kinematic analyses of
reaching have shown that patients may use a similar pattern of movement to
able-bodied controls but need more time to determine the spatial location of
the target (Fisk and Goodale 1988). Hemispatial neglect can be a subtle deficit,
showing up only in the temporal and spatial form of movement detected on
kinematic analysis (Goodale et al. 1990).
Neglect is most commonly seen in visually guided activities. The individual with
a spatial impairment may not dress or wash the left side of the body or shave
on the left (autotopagnosia), eat food only on the right side of the plate, brush
teeth on one side only, start reading a sentence from the middle of the page and
fail to attend to objects and people on the left side. Walking through a door
way, the individual may veer to one side or bump into the door frame. The per
son may deny the left arm or leg belongs to them, or deny the existence of
hemiplegia (anosognosia). Difficulties with writing, reading and drawing also
come under this general heading, together with impairments of line orientation
and verticality perception. Clinical features regarded as making up the syn
drome of neglect may, however, have nothing more in common than the fact
that there is involvement of the left side of space (Halligan and Marshall 1992).
It has been proposed that the inferior and posterior parietal cortex may be crit
ically involved in monitoring and directing attention within the visual environ
ment, with more anterior lesions resulting in impaired initiation and execution
of movements within the contralateral side of space (Coslett et al. 1990).
However, neglect phenomena cannot be explained in terms of a single mech
anism (Halligan and Marshall 1992, Riddoch and Humphrey 1994) since a
brain lesion, even a relatively discrete one, disrupts an entire functional system.
Three main hypotheses have emerged in attempts to explain the impairment
(Ladavas et al. 1994):
a representational hypothesis which considers the impairment to be an
inability to form a whole representation of space (Bisiach and Vellor 1988)
an arousal hypothesis in which the impairment is viewed as a selective
loss of orienting responses to space (Heilman et al. 1987)
a selective attention hypothesis in which there is an imbalance in the
attentional system which favours shifts to the opposite side
(Ladavas et al. 1990).
Attention to one side to the exclusion of the other seems very compelling to the
affected individual. For example, stimuli in the right visual field may provide a
strong magnet, making it difficult to disengage attention once it is visually cap
tured (Posner et al. 1987). For patients to be able to shift visual attention they
need to be able to disengage their attention from one target, shift it to another,
then focus attention on this new target (Herman 1992). A patient may report
difficulty turning the eyes toward the left side without also moving the head.
Different components of attention may be affected (Ladavas et al. 1989,
Humphreys and Riddoch 1992):
arousal-activation mechanisms, the physiological readiness to respond to
stimuli wherever they are located (Heilman and van den Abell 1980)
orientation and selective attention to a space (Posner et al. 1987)
shifting attention from one target to another (Mark et al. 1988)
intention, the motor activation necessary to respond to a stimulus
(Verfaelli et al. 1988).
Outcome for individuals with neglect may be affected by the state of the individ
ual's brain pre-stroke. In a small study of patients with large right hemisphere
lesions, the degree of recovery was inversely related to the amount of pre-stroke
cortical atrophy, i.e. recovery was greatest in those whose pre-stroke atrophy
was least (Calvanio et al. 1993).
The term 'pusher syndrome' is used by therapists (Davies 1985, Ashburn
1995) in an attempt to explain a puzzling behaviour in some individuals who
lean toward the hemiplegic side, resisting strongly any attempt at passive cor
rection of posture that would move weight toward or across the midline of the
body to the non-affected side. There has been little investigation of this phe
nomenon and it is not certain whether the behaviour observed by therapists
represents a separate syndrome or is an adaptive or learned behaviour in a per
son who has some of the impairments described above under unilateral spatial
A Danish study (Pedersen et al. 1996) examined the incidence of ipsilateral
pushing in an attempt to establish whether or not there is such a syndrome and
to determine the effect of pushing on functional recovery. The incidence was
found to be 10% of the 327 subjects studied and the diagnosis of 'pusher' was
provided by therapist observation and not by any standardized or reliable meas
ure. All patients with unilateral pushing had the more severe strokes, were no
more likely to suffer neglect or anosognosia than the non-pushers, were distrib
uted evenly between left- and right-sided lesions, and spent longer in rehabilita
tion. The authors concluded that there was, therefore, no support for the
so-called 'pusher syndrome', in the sense of a syndrome encompassing both
physical and neuropsychological symptoms.
There is some evidence that individuals with this problem have an impairment
in verticality perception. Misperception of verticality has been reported to be
associated with falling in elderly individuals, including those with stroke
(Tobis et al. 1981). Lateropulsion has been described by Dieterich and Brandt
(1992) as part of Wallenberg's syndrome, which includes poor visual percep
tion of the vertical. It has been hypothesized that the midline of the trunk
and/or head serves to divide space into a right and a left sector, i.e. the body's
midline serves as a physical anchor for the calculation of internal egocentric
coordinates for representing the position of the body relative to external
objects (Karnath et al. 1991). The relationship with verticality perception
needs further investigation.
Anecdotally, the tendency to lean toward the affected side and to resist
strongly attempts by the therapist to shift the person toward the other side
seems to disappear after a short period of active and intensive balance training,
and as patients regain their ability to move about voluntarily in sitting and
standing. If patients have experienced a change in their perception of the midline
of the body, this would explain the anxiety and fear expressed if they are moved
passively by the therapist (e.g. into a more 'symmetrical' sitting position). It is
possible that some passive techniques used in therapy, which involve the
patient being moved about by the physiotherapist rather than moving actively,
may reinforce the abnormal behaviour and cause it to persist for longer. If an
individual's subjective midline has shifted to one side, being moved by another
person to the opposite side could be perceived as a large excursion away from
the midline. Affected patients are clearly distressed when attempts are made to
move them to the unaffected side since they seem to think their appropriate
position in space is to the left of true centre (Bohannon 1996). They appear to
cope much better when they practise tasks such as reaching for an object
which requires active movement of the body to both sides, with visual cueing
and feedback provided by the success or otherwise of their performance. It can
be helpful to point out that they are being misled about the nature of vertical
and to encourage them to look at objects they 'know' to be vertical (the leg of
a bed, a door frame); i.e. to intellectualize the 'idea' of vertical and to try to
align themselves with known vertically oriented objects. There is some evi
dence that this approach involving active practice of specific tasks can be effec
tive (Bohannon 1998).
In conclusion, it is evident that somatosensory impairments can improve sig
nificantly following specific sensory retraining. However, there is little evidence
of carryover from methods which are not aligned to meaningful tasks. It is
likely that specific sensations are improved by the natural ongoing sensory
inputs occurring during active training and practice of tasks. Task-oriented
training provides the opportunity for the lesioned system to put together sen
sory inputs and appropriate motor outputs. Neuropsychological methods of
intervention include repetitive visual scanning, cueing and sustained attention
For training to be rationally planned, sensation must be tested. Although many
of the tests used in the clinic are subjective, recently tests have been developed
which, if performed under standardized conditions and with the appropriate
Box 6.4 Summary of sensory and perceptual-cognitive impairments
Somatosensory impairments include partial or complete loss of discrete sensa
tions, including impairment of joint position and movement sense and tactile
Visuocognitive impairments include visual field loss and visuospatial agnosia
Perceptual-cognitive impairments include disorders of body image and
impaired awareness of one side of space (usually the L) as in unilateral spatial
agnosia (neglect)
Lateropulsion (the 'pusher syndrome') may be associated with a disorder of
verticality perception
Therapists should avoid moving passively a person who leans persistently to
one side but instead should encourage active balancing exercises
Specific sensory retraining appears effective in improving sensory discrimin
ation but carryover into improved function needs to be investigated
equipment, enable a quantification of the individual's ability to perceive spe
cific sensory inputs (Yekutiel and Guttman 1993, Carey 1995, Winward et al.
1999). Where facilities are available, afferent projections from limbs to cere
bral cortex are studied using somatosensory evoked potentials (Burke and
Gandevia 1988, Kuzoffsky 1990). Perceptual-eognitive impairments are tested
by a clinical neuropsychologist.
Sensory and perceptual-cognitive impairments are summarized in Box 6.4.
Strength training and
physical conditioning
The major goal of physiotherapy in neurological rehabilitation is the optimiza
tion of functional motor performance. Major impairments limiting motor per
formance are muscle weakness or paralysis, soft tissue contracture, lack of
endurance and physical fitness.
This appendix provides some general guidelines for strength training in rela
tion to the optimization of motor control and skill, training to increase phys
ical fitness and endurance, and methods of decreasing stiffness and preserving
soft tissue length.
The physiological factors which affect strength are structural and functional.
Structural factors include the cross-sectional area of the muscle (its size), the
density of muscle fibres per unit cross-sectional area and the efficiency of
mechanical leverage across joints. The functional factors include the number,
type and frequency of motor units recruited during a contraction, the initial
length of muscles and the efficiency of cooperation between synergic muscles
involved in the action. In addition, biomechanical factors also affect strength.
For example, the viscoelastic properties of the contractile and non-contractile
tissues absorb energy when an active muscle is stretched and this can be used
to increase the concentric force of the contraction. Strength is therefore a
function of the properties of muscle and depends on intact neurological func
tion (Buchner and de Lateur 1991).
It follows that strength training is necessary after stroke to improve the force
generating capacity and efficiency of weak muscles and to improve functional
motor performance. The relationship between strength and performance is
complex. The amount of strength required depends on the person's physical
size (weight of hand, total body weight) - i.e. relative as opposed to absolute
strength (Buchner and de Lateur 1991) - and on the action to be performed
(walking on a flat surface vs stair climbing). Skilled motor performance, how
ever, requires the following:
each muscle involved in the action has to generate peak force at the length
appropriate to the action
this force has to be graded and timed so synergic muscle activity is
controlled for task and context
the force has to be sustained over a sufficient period of time
peak forces must be generated fast enough to meet environmental and task
demands such as increasing walking speed to cross the road at traffic
Although reductions in muscle force production and its control cause major
motor dysfunction after a lesion of the neuromotor system, strength training
has been eliminated from many therapy programmes over the past five decades,
to a large extent due to the continuing influence of certain physiotherapy
beliefs. First, weakness has been thought to be due to inhibition from spastic
antagonists and 'low tone' of agonists rather than a direct result of the reduc
tion in descending inputs to spinal motoneurons; second, exercises causing
effort are therefore contraindicated as they are expected to increase spasticity,
co-contraction and abnormal movement patterns.
These views are still being taught and many clinicians are therefore maintain
ing an approach to rehabilitation that is no longer supported either by scien
tific theory or clinical evidence. Clinical research has shown that effort applied
in strength training does not increase spasticity (reflex hyperactivity), associ
ated movements, co-contraction or resistance to passive movement" (Butefisch
et al. 1995, Davies et al. 1996, Sharp and Brouwer 1997, Brown and Kautz
1998, Smith et al. 1999, Teixeira-Salmela et al. 1999, Ada and O'Dwyer 2001,
Bateman et al. 2001). On the other hand, not only does strength training result
in increased muscle strength following stroke, but also improved functional
performance and decreased spasticity, for example decreases in resistance to pas
sive movement, stretch reflex hyperactivity and co-contraction (Butefisch et al.
1995, Miller and Light 1997, Smith et al. 1999, Teixeira-Salmela et al. 1999).
The dynamic stretching that occurs during active exercise may playa part in
decreasing reflex hyperexcitability and increasing muscle compliance (Otis et al.
1985, Hummelsheim et al. 1994). The evidence is mounting that repetitive,
task-oriented strength training is effective following stroke, as is also being
*Variously tested by resistance to passive movement on the Ashworth scale, response to
pendulum test, presence of abnormal movement patterns, or inappropriate muscle activity or
co-contraction on EMG.
shown following cerebral palsy and traumatic brain injury. It is also likely that
practice and training, by improving aspects of motor control such as timing
of muscle activations, would result in decreased co-contraction (Carr and
Shepherd 1987). There seems little doubt that active exercise and training pro
voke efficacious changes in motor control.
There is another consideration when planning intervention. Many individuals
who have stroke are elderly. For some, their pre-stroke physical condition may
have been poor due to ill health, excessive weight, or a sedentary lifestyle.
Consequently, reductions in muscle strength due to the lesion, together with
pre-existing reductions in joint flexibility, strength, endurance and physical fit
ness, combine to impact severely on the patient's ability to regain the necessary
functional motor skills. Poor physical condition and being elderly are not bar
riers to strength and fitness training post-stroke but are instead imperatives. It is
known that regular moderately vigorous exercise has positive effects on muscle
mass and strength, postural stability and prevention of falls, joint flexibility,
bone density, psychological function and cardiovascular responses to exercise
(read Mazzeo et al. 1998 for review and recommendations). The avoidance of
strength training in patients with muscle weakness who are also lacking in fit
ness is very likely, therefore, to have a significant negative impact on functional
outcome after stroke (Miller and Light 1997).
Factors associated with the nature of the lesion, the patient's pre-stroke phys
ical condition and subsequent physical inactivity provide, therefore, several
compelling reasons for emphasizing strength training following a stroke, sum
marized as follows.
1. Muscle weakness is a major impairment to effective functional performance
a stroke usually results in some degree of muscle weakness, including
paralysis, primarily as a direct result of a reduction in descending inputs
on spinal motoneurons and of the number of motor units activated
the immobility which ensues results in mechanical and functional
changes to the muscles and connective tissue, and predisposes to further
reductions in strength
elderly individuals may have had varying degrees of muscle weakness
and reduced endurance and cardiovascular responses pre-stroke due to
a decline in physical activity.
2. There is mounting evidence that strength training is effective following
muscle strength is increased
increased strength can be associated with improved functional
when incorporated into an intensive activity programme, exercise
capacity and endurance are also improved.
3. There is no evidence that spasticity (hyperreflexia) or hypertonus (resistance
to passive movement) increase, and some evidence that they decrease, after
strength training.
The mechanisms underlying the improvement of muscle strength and motor
control reported following stroke may be similar to those after strength training
in able-bodied individuals. Neuromuscular adaptations associated with muscle
strengthening and task acquisition are described by several authors (e.g. Sale
1987, Rutherford 1988, Moritani 1993). It is generally accepted that a large
part of any early increase in muscle strength in the able-bodied is due to neural
adaptations, including task-specific adaptations to neural drive, which are both
quantitative (increased neural drive to target muscles) and qualitative (reduced
co-contraction and improved coordination among synergists) (Hakkinen and
Komi 1983, Rutherford and Jones 1986, Carolan and Cafarelli 1992, Yue and
Cole 1992). Muscle hypertrophy or an increase in cross-sectional area plays a
part in later increases in strength. Furthermore, neural adaptations appear to be
specific to the training action or task. Strength training effects are, therefore,
both neural and structural, and include:
stimulation of neuromuscular changes due to reorganized drive from
supraspinal centres. Changes include enhanced muscle excitation,
improvements in recruitment of motoneuron pool, inhibition of antagonist
muscles, motor unit activation and synchronization of the firing pattern of
motor units (Hakkinen and Komi 1983, Sale 1987)
stimulation of metabolic, mechanical and structural muscle fibre changes
that result in larger and stronger muscles due to an increase in actin and
myosin protein filaments (Sale 1987, Thepaut-Mathieu et al. 1988).
These effects are reported from studies of able-bodied individuals but their rele
vance to individuals with motor impairments is obvious. It has been shown with
able-bodied young men on an 8-week progressive strength training programme
that an increase in maximal force produced by the lower limb extensors was
accompanied by increases in neural activation (integrated EMC) of those
muscles (Hakkinen and Komi 1993). It is likely that increased strength following
stroke also results from increased mechanical efficiency of the muscles them
selves, improved limb control and motor learning (Rutherford and Jones 1986,
Rutherford 1988). Increased strength occurs in muscles not directly strength
ened but which, since they act as stabilizers and synergists during the exercise,
are also affected positively (Fiatarone et al. 1990).
There may be some advantageous effect of increase in muscle size after stroke
(Hakkinen et al. 1998, Harridge et al. 1999), since it is likely that muscle atro
phy can become a factor in disuse weakness after stroke as it is during other con
ditions of imposed limb immobility. In animal experiments, quadriceps atrophy
has been reported as early as 3 days after immobilization (Lindboe and Platou
1984) with 30% reduction in cross-sectional area within 1 month (Halkjaer
Kristensen and Ingemann-Hansen 1985). One study has reported that muscle
weakness and atrophy due to disuse are essentially preventable by exercise that
activates high threshold motor units such as sit-to-stand, stair walking and eccen
tric knee extension training (Hachisuka et al. 1997). In the very elderly (85-97
years) an increase in quadriceps cross-sectional area, measured at mid-thigh with
magnetic resonance imaging, has been demonstrated after 12 weeks of isotonic
knee extensor strength training without any change in muscle activation. The
authors suggest that increased muscle mass may have functional and metabolic
benefits for the very elderly (Frontera et al. 1988, Harridge et al. 1999).
Some of the improvements in activity levels reported after strength trammg
may also be due to improved psychological state, self-concept (McAuley and
FIGURE 7.1 (Left) Hip, knee and ankle muscles contract in synchrony in kinetic chain exercises
such as step-ups. (Right) An exercise for quadriceps femoris that isolates movement
to the knee joint.

Rudolph 1995) andJor energy levels. The effects of 12 months of bicycle ergo
meter training included improved self-concept and functional status in chronic
stroke (Brinkman and Hoskins 1979). Improvements in functional status are also
attributed to positive feedback from success in training, and social enjoyment.
Type of exercise
There are several methods of strength training in use in rehabilitation:
Isometric exercise: a muscle contraction is performed with little movement
or change in muscle length
Isotonic exercise: an exercise in which the resistance remains constant;
includes active gravity-eliminated exercise, free weights *, weight-loaded
machines and elastic band exercise
Isokinetic exercise: exercise that uses an isokinetic dynamometer in which the
speed is kept constant and the variable resistance matches the force applied.
Single joint and kinetic chain exercises
Isotonic and isokinetic exercises may be performed by isolating one segment
(sometimes called open-chain exercise). In these exercises, the limb moves
against resistance with the distal segment free, as in the seated knee extension
exercise in Figure 7.1. Functionally, however, in many everyday actions, the foot,
shank and thigh behave as a linked segmental chain (Fig. 7.3). The term lower
limb kinetic chain (or 'closed-chain') exercise (Palmitier et al. 1991) is used to
describe exercises that recruit all three links together. Movement at one joint,
of necessity, produces movement of the other joints.
"Free weights are not purely isotonic since muscle tension varies according to gravity and
segmental relationships.

~ 1
\ /

238 Appendices
FIGURE 7.2 (a) Modified leg press exercise (affected leg) performed independently on a tilt
table. He uses a counter to keep track of number of repetitions. (b) Leg press
exercises for weak right leg using pneumatic strength training equipment. Note that
a more reclined position can be used to increase the range of hip extension.
(Courtesy of Keiser Sports Health Equipment Inc., Fresno, CA.)
(a) (b)
Kinetic chain weightbearing exercises are typically given to strengthen lower
limb extensor muscles using body weight and other forms of resistance, and
include exercises in which the body mass is lowered and raised over the feet,
for example, step-ups (see Fig. 3.22), modified squat to stand (see Fig. 2.17),
heels raise and lower (see Fig. 3.24), and leg press exercises (Fig. 7.2).
These exercises take advantage of the specificity principle in that muscles are
exercised concentrically and eccentrically in a movement pattern that shares
some of the dynamic characteristics of commonly performed motor actions such
as sit-to-stand, bending down to pick up objects in standing, stair climbing and
descent (Palmitier et al. 1991) (Fig. 7.3). Exercises such as these not only enable
Strength training and physical conditioning 239
FIGURE 7.3 Diagram showing lower limb segments and eight muscle groups affecting segmental
rotation during f'exion and extension of the lower limbs over a fixed foot. g max,
maximum gravity; gastroc, gastrocnemius; hamstr, hamstrings; rect fem, rectus
femoris; tib ant, tibialis anterior; tib post, tibialis posterior. (Reprinted from Journal
of Biomechanics, Vol. 26, AD Kuo and FE Zajac, A biomechanical analysis of muscle
strength as a limiting factor in standing posture, p. 139, copyright 1993, with
permission from Elsevier Science.)
practice of lengthening and shortening contractions of muscles spanning several
joints (e.g. biceps femoris, gastrocnemius), they also have the advantage of
training the complex neural control involved in these actions (Pinniger et al.
2000), and place functional stresses on the limb which are similar to those in
many weightbearing activities (Shelbourne and Nitz 1992). It is apparent on
clinical observation that individuals can have difficulty after stroke with actions
that involve switching from concentric to eccentric muscle activity, particularly
when weightbearing. This has been observed also in elderly individuals.
Exercises involving actions that require a switch from one mode of contraction
to the other, such as step-ups, can provide a focus for training coordination
while also providing an action-specific strength training stimulus.
In weightbearing kinetic chain exercises, synergistic muscle activity plays a large
part in the action. Consequently, the individual is not only increasing the
strength of a group of muscles but is being trained to control the muscle forces
produced across the segmental linkage. There is some clinical evidence that
weightbearing exercises for the lower limb can improve strength and functional
performance in individuals with disability. For example, a programme of this type
of strength training was followed by improved isometric strength, gait and sit
to-stand (STS) performance in children with cerebral palsy (Blundell et a1. 2002),
and practising step-up exercises was associated with improved gait after stroke
and in elderly subjects (Nugent et a1. 1994, Sherrington and Lord 1997).
It is likely that in patients with very weak muscles - say, Motricity Index (MI)
score < 3 - any method of strength training, including exercises performed
with the distal segment free, may be of value. There is some theoretical support
for the view that, in this situation, any exercise that stimulates muscle activa
tion and requires the muscle to work eccentrically, concentrically or isometr
ically will transfer into improvements in functional motor performance.
Isokinetic exercise, where available, may be particularly useful in patients who
have difficulty sustaining appropriate levels of muscle activation through
range. However, when muscles are stronger, i.e. above a certain threshold
(unknown at this time but let's say MI score> 3), that strength training
should consist of exercises that are dynamically similar to the actions being
trained in order to train motor control and maximize carryover into function.
Exercises are progressed by increasing the number of repetitions, the range of
movement (e.g. height of step) and the amount of resistance or load.
Concentric and eccentric exercise
The force-producing capacity (tension regulation) of muscle differs during
lengthening and shortening contractions (Pinniger et a1. 2000). Voluntary
eccentric contractions produce greater muscle force than concentric muscle
contractions (Komi 1973) yet appear to involve lower rates of motor unit dis
charge than concentric contractions (Tax et a1. 1990), i.e. lesser levels of mus
cle activation are needed for greater levels of force (Westing et a1. 1990).
Eccentric exercise has greater mechanical efficiency (Komi and Buskirk 1972).
At comparable work loads, the cost in terms of metabolic energy is less than in
concentric exercise (Asmussen 1953).
In an able-bodied population, utilizing both concentric and eccentric muscle
contractions in strength training has been shown to produce better gains in
strength than concentric contractions alone (Hakkinen and Komi 1981). When
the muscle is actively stretched in an eccentric contraction, tension in the series
elastic component increases and the stored elastic energy is used in the subse
quent concentric action (Asmussen and Bonde-Petersen 1974, Svantesson et a1.
1999). In patients with neuromotor impairment, it is possible that, at the neural
level, concentric activation of weak muscles may be facilitated by the enhanced
muscle spindle activity occurring as a result of the rapid switching from eccen
tric to concentric muscle activity (Burke et a1. 1978). It is known from studies of
the able bodied that, if an eccentric muscle contraction immediately precedes a
concentric contraction, as in the brief flexion counter-movement of the lower
limbs which immediately precedes the extension phase of the vertical jump, the
concentric phase generates more force due to the effect of the stretch-shortening
cycle (Asmussen and Bonde-Petersen 1974, Bosco et al. 1982, Svantesson et al.
1994). This mechanism seems to playa significant role in many everyday actions
including sit-to-stand (Shepherd and Gentile 1994) and walking (Komi 1986).
Elastic band resistance exercise
Elastic bands provide an inexpensive and simple means of exercising which can
be carried out by patients on their own. Examples of such exercises are given in
the chapters on walking and upper limb function. The bands are provided in a
range of resistance levels which enables prescription to suit an individual's
strength level. Tests of these bands have provided details of the approximate
amount of resistance provided by the particular colours. For example, for
Theraband* at 100% deformation: yellow = 13 N, green = 25 N, black =
36N (Simoneau et al. 2001). There are several other colours, however, and
recent work has shown that the red band provides resistance between the yel
low and green bands. The silver band provides the largest amount of resistance,
approximately double that of the green band (Patterson et al. 2001).
Each exercise should start with the slack of the elastic band taken out. One
advantage of the elastic band over hand weights is its ability to provide variable
resistance throughout the range of movement. Simoneau and colleagues (2001)
describe the properties of elastic bands and factors to consider in their use.
There is some difficulty in controlling the exact amount of resistance offered by
elastic bands (Hintermeister et al. 1998). Nevertheless, for individuals following
stroke the use of these bands is recommended as a means of increasing muscle
strength, preserving or increasing joint range and muscle extensibility, and
encouraging unsupervised exercise for both upper and lower limbs. Progression
involves increasing intensity of exercise (number of repetitions), load (using
band colour to increase resistance) and frequency (number of times per day).
There is evidence that exercises using elastic band resistance to isotonic move
ments can be effective in increasing strength in elderly individuals (Mikesky
et al. 1994, Skelton et al. 1995), with some carryover into improved balance and
gait (Topp et al. 1993, 1996). It is yet to be investigated, however, whether or
not muscle strength is increased by such exercise in individuals following
stroke, or if there are other benefits such as improved motor performance,
increased activity levels and, in the case of the upper limb, a decreased ten
dency to develop adhesive capsulitis with painful shoulder.
Specificity of strength training
The conventional approach to atrophy and muscle weakness in individuals fol
lowing musculoskeletal trauma is to increase strength and restore muscle bulk
by high resistance strength training exercise (Rutherford 1988). The assumption
is often made that improved strength will transfer automatically into improved
"Theraband, The Hygenic Corporation, Akron, Ohio.
functional ability. This may not, however, be the case. The body adapts specific
ally to the demands imposed upon it, and research over many years has shown
that exercise effects tend to be specific to task and context (Rutherford 1988,
Morrissey et al. 1995), with the greatest changes occurring in the training exer
cise itself. For example, increases in strength of a muscle appear to be specific to
such factors as joint angle (muscle length), subject's posture (standing or
supine), velocity of movement and muscle contraction type. Muscle length
specificity is of particular interest, with the greatest increases in maximum vol
untary contraction consistently found at the training angle after isometric exer
cise. Results of one study showed that the shorter the muscle length at which
training was carried out the more the gain in maximum voluntary contraction
was limited to the training angle (Thepaut-Mathieu et al. 1988).
The principle of specificity is usually explained with reference to such factors as
the structure and function of particular muscles, biomechanical constraints such
as length of moment arm, and the nature of synergic cooperation. Functional
actions are made up of complex movements of the multisegment linkage which
require strength, coordination and balance. The relationship between muscles
that span one joint and those that span two (or many) joints is particularly
complex and little understood (Bobbert and van Soest 2000).
The extent to which the intrinsic strength of stabilizer and other synergic muscles
influences the strength of a prime mover muscle has been little investigated.
However, a recent analysis of a task that involved pushing a heavy object
(Kornecki et al. 2001) drew attention to the importance of muscles acting as sta
bilizers to 'stiffen' the degrees of freedom not involved in the action. The study
found that the ability to stabilize a key joint influenced muscle activation patterns.
Stabilizing the wrist joint was associated with decreased EMG in the upper arm
muscles (triceps brachii, deltoid), the prime movers for the pushing movement.
After stroke, when muscle weakness is part of a more general motor control
impairment, it may be critical that strength training is oriented toward specific
tasks and designed to ensure the greatest transfer possible into improved func
tional motor performance. Strength training may need to focus on exercising
muscles through a specific part of the range (e.g. calf muscles from fully length
ened to mid-range for stance phase of gait). It is therefore critical that a close
examination is made of the patient's ability to generate and sustain force as
active motor tasks are practised, and that this is followed up by exercise and
training specifically targeting the muscles in which weakness is evident and
apparently affecting function.
Transfer from strength to function
It is intuitive that a relationship should exist between strength and function;
however, the nature of the relationship appears complex and dependent on the
degree of muscle weakness. An hypothesis put forward by Buchner and de
Lateur (1991) suggests that the relationship between strength and function is
curvilinear and activity specific. In a study of elderly individuals, Buchner and
colleagues (Buchner et al. 1996) showed that a curvilinear relationship exists
between lower limb strength and walking velocity. They pointed out that this
relationship reflects a mechanism by which small changes in strength may pro
duce relatively large changes in performance in very weak adults, while large
changes may have little or no effect in able-bodied adults. When muscles are
beyond a certain task-dependent threshold (a threshold that for most everyday
tasks is well below normal strength values), exercises are required that are task
oriented. Buchner's hypothesis may explain the discrepant results in studies
reporting a relationship between strength and function.
A recent study illustrated the beneficial transfer effects of task-oriented strength
training in an able-bodied group. Glenohumeral internal and external rotation
exercises (incorporating elastic band and hand weight resistance) performed by
tennis players in standing, with shoulder at 90 abduction, resulted in signifi
cant increases in force production of the actions practised, with a transfer into
increased speed of serve (Treiber et al. 1998). Transfer is unlikely to occur,
however, unless subjects are also practising the task to be learned (in the above
case, the tennis serve), even in some modified context.
Transfer from single joint exercise to kinetic (weightbearing) actions may be
limited except where muscles are very weak. It has been shown that open
chain knee extension exercise can increase strength of quadriceps in terms of
the weight that can be lifted through range (the exercise), with no improve
ment in maximum power output measured in isokinetic cycling (a closed-chain
action in the pedal-pressing phase). Some studies have produced more equivo
cal results. This may be due to the relevance of the exercises to the functional
action being learned or to the individual's strength level.
Variations in body position also affect muscle activity patterns. An early experi
ment showed that strength training of elbow muscles carried out in standing
resulted in greater increases in strength than the same exercises carried out in
supine (Rasch and Morehouse 1957). The strength increases found in standing
were in large part the result of learning to coordinate all the muscles involved
in the movement, which included those controlling balance.
Strength training that involves repetitive practice of the action being learned can
improve strength and endurance if load is progressively increased in some way.
For example, walking on a treadmill (and overground) can be progressed by
increasing speed and slope; STS can be practised repetitively from increasingly
lower seat heights, and at increasing speed. If done repetitively with sufficient
load, such training provides a means of strengthening lower limb muscles and
increasing endurance, as well as control over the dynamics of the particular
action to be regained.
Also critical to the regaining of effective performance is the development of
flexibility of performance, which is achieved by practising the action under a
variety of different conditions, i.e. in different environmental and task con
texts. Repetitive practice of the action to be learned can therefore have dual
benefits, enabling the patient to practise the action as well as increasing muscle
strength (Rutherford 1988). Strength training for individuals following stroke
should be considered to have as its main effects improved performance in
actions required in daily life and prevention of falls.
It may follow that when individuals have a marked degree of weakness, i.e.
below a certain threshold of strength relative to a particular activity such as
walking, several types of strength training may be effective. Any increase in
strength is likely to result in some improvement, such as the ability to at least
attempt walking: i.e. any exercise that requires a muscle to contract and
lengthen its fibres against even a small amount of resistance is likely to be bene
ficial initially when muscles are very weak. Above a particular activity-specific
threshold of strength, however, strengthening exercises may have minimal effect
if they do not share similar dynamic characteristics to the action being trained.
Eliciting activity in very weak muscles
Following stroke, exercises to strengthen muscles are selected according to the
degree of weakness. An estimation of muscle strength is gained by simple testing
using the Medical Research Council (MRC) grades or the Motricity Index (MI)
which is based on the MRC grades. If muscles can be voluntarily activated but
are very weak (say, grade 2-3), strength training can include partial body
weight resistance, or resistance through a small range of movement (e.g. STS
from a high seat), lifting small weights through a limited range, elastic band
exercises, and concentric and eccentric exercise on an isokinetic dynamometer.
Training may require an exploration to find the conditions under which a
very weak or paralysed muscle (grade 0-1) can be stimulated to contract.
Figure 5.14a shows a method of eliciting eccentric activity in triceps surae in
an individual who had severe weakness of this muscle. Since lower levels of
muscle activation are required for the same force effect in eccentric compared
to concentric exercise, attempts at eccentric contraction may enable an individ
ual with very weak muscles to improve activation of those muscles. For exam
ple, if a person is unable to contract quadriceps, a voluntary contraction may
be elicited eccentrically by attempting to lower the shank slowly with an
isokinetic dynamometer (see Fig. 3.26).
Biofeedback may be combined with exercise to spur the individual on to
greater efforts (Wolf et at. 1994). Mental practice may also be useful in some
individuals (Page et at. 2001). Early after stroke, electrical stimulation (ES) or
electromyography-triggered ES may give the person the idea of muscle activation
if a voluntary contraction appears impossible. It has been suggested that repeti
tive motor training via neuromuscular ES may facilitate motor learning by modi
fying the excitability of specific motor neurons (Chae and Yu 1999). There is
some evidence that ES can have a positive effect on joint movement, muscle
strength and functional performance (Baker et at. 1979, Lieber 1988, Dimitrijevic
et al. 1996, Glanz et at. 1996, Chae et at. 1998, Sonde et at. 1998). These meth
ods are discussed further in the context of upper limb training (Ch. 5).
Efficacy of muscle strength training after stroke
Information on physiological responses to exercise in an individual with a
brain lesion is limited (Lexell 2000). However, there is a rapidly growing body
of knowledge about the effects of exercise on strength and function in the
elderly, and following stroke. It is increasingly evident that muscle weakness
after stroke is modifiable through strength training of an appropriate intensity,
even in very elderly individuals (Sharp and Brouwer 1997, Mazzeo et a1. 1998,
Teixeira-Salmela et a1. 1999, Ng and Shepherd 2000).
Studies have reported the following changes after periods of strength training
and physical conditioning:
Increases in muscle strength, improved postural stability and reduction in
falls in the elderly (Aniansson et a1. 1980, Aniasson and Gustafsson 1981,
Sauvage et a1. 1992, Fiatarone et a1. 1990, 1994, Judge et a1. 1993, Tinetti
et a1. 1994, Campbell et a1. 1997, Gardner et a1. 2000)
Increases in muscle strength after stroke (Sunderland et a1. 1992, Engardt
et a1. 1995, Sharp and Brouwer 1997, Sherrington and Lord 1997, Brown
and Kautz 1998, Duncan et a1. 1998, Teixeira-Salmela et a1. 1999,2001,
Weiss et a1. 2000).
Many of these studies also document a positive relationship between increased
strength and functional performance. For example, increased strength of lower
limb muscles (hip and knee flexors and extensors, ankle dorsi- and plantarflex
ors) is associated with improvements in aspects of:
gait performance (Nakamura et a1. 1985, Bohannon and Andrews 1990,
Nugent et a1. 1994, Lindmark and Hamrin 1995, Sharp and Brouwer
1997, Krebs et a1. 1998, Teixeira-Salmela et a1. 1999,2001, Weiss et a1.
ability to balance (Hamrin et a1. 1982, Weiss et a1. 2000)
stair climbing (Bohannon and Walsh 1991).
A single case study of static dynometric strength training with computer-driven
visual feedback of force showed an increase in the amount of static direction
specific force produced through the lower limb plus improvement on the Timed
'Up and Go' test, comfortable gait speed and endurance (Mercier et a1. 1999).
Furthermore, a significant level of improvement was maintained 6 weeks later.
This study was designed to test a novel method of repetitive resisted exercise in
which the subject was required to control the direction of force, responding to
feedback on both direction and amplitude.
In another study of lower limb strength training, increases in quadriceps and
hamstring muscle strength were accompanied by increases in gait velocity fol
lowing a 6-week programme of exercise which included stretches and isokin
etic strengthening for quadriceps and hamstrings (Sharp and Brouwer 1997).
A 10-week programme of strength training for hip, knee and ankle muscles
plus physical conditioning exercises resulted in gains in the Human Activity
Profile and the Nottingham Health Profile, both of which reflect the ability to
perform household tasks and recreational activities, and increases in gait speed
and performance. After training, patients were able to increase power gener
ation by ankle plantarflexors and hip flexors and extensors (Teixeira-Salmela
et a1. 2001). In another study, 15 patients increased force output in their
paretic lower limb after a period of pedaling against various loads (Brown and
Kautz 1998).
The role of peak muscle power (the product of force and velocity of muscle
shortening) in functional independence has been examined in elderly individuals
(Bassey et al. 1997), with the suggestion that power may be more directly related
to impaired physical performance than strength in the elderly. This is of interest
in neurorehabilitation, given the reported deficits in power production reported
after stroke (Olney and Richards 1996). In the elderly, the velocity with which
force can be generated declines along with muscle strength itself, power decreas
ing to a greater degree than strength. Recent studies have reported that peak
lower limb muscle power, tested on a leg press device is a predictor of self
reported functional status in older women (Foldvari et al. 2000), and there is
some evidence that lower limb strength and power can be improved by high
velocity power training in this population (Fielding et al. 2002).
Treadmill exercise can increase muscle strength and endurance. A recent study
of task-oriented aerobic treadmill training carried out three times a week for
3 months showed increased strength of quadriceps measured on an isokinetic
dynamometer (Smith et al. 1998).
There are few studies of strength training for upper limb muscles after stroke.
However, a strength training programme involving repetitive hand and finger
flexion exercises performed against various loads showed increases in strength
of hand grip, wrist extension strength and speed of wrist extension (Butefisch
et al. 1995). Another study reported a relationship between increased grip
strength and improved hand function (Sunderland et al. 1989).
Whether or not increases in muscle strength and function in individuals with
stroke are maintained after a training programme has concluded is not clear.
Nevertheless, it is likely that the opportunity for long-term, periodic exercise
training at home and in community classes, and attendance at a specialized
gymnasium, is necessary for many individuals after stroke.
Exercise prescription
There are certain basic considerations in planning a muscle strengthening pro
gramme. The exercise dosage can be increased by increasing the number of
repetitions, the number of sets, and the resistance provided. In general, in able
bodied subjects, muscles should be exercised to the point of fatigue but not
pain in order to obtain some change (Wajswelner and Webb 1995). Patients
should be warned that they may experience a small degree of delayed muscle
soreness. The exercise should be as specific as possible to the functional actions
being trained to ensure carryover (Morrissey et al. 1995). In the case of
patients with very weak muscles, however, any type of exercise which results
in the generation of some force can be practised.
Strength trammg in the able-bodied appears to produce greatest strength
increases when subjects train at high but submaximalloads. As a general guide,
the patient should attempt 10 repetitions of 80% of the maximum possible load
that can be lifted in one attempt. Ten repetitions is called the 10 repetition max
imum (10 RM). This is set as a goal, although a very weak person may start at
50% of the maximum load and with fewer than 10 repetitions (Fiatarone et a1.
1990, Teixeira-Salmela et a1. 1999). It should be noted that the repetition max
imum is the maximum number of repetitions that can be performed maintaining
good form (Wajswelner and Webb 1995). The weight or load for a given RM
(in this case 10) is the load that can be lifted 10 times without gross error.
A recent study provides guidelines for exercise prescription. Teixeira-Salmela
and colleagues (1999) investigated the effects of a strengthening and physical
conditioning programme in 13 individuals with chronic stroke. Each session
consisted of:
5-10min warm-up (calisthenics, stretching, range of motion exercises)
aerobic exercises (graded walking, stepping or cycling)
strength training (hip, knee and ankle flexors and extensors)
5-10 min cool-down (muscle relaxation, stretching).
Strength training consisted of isometric, concentric and eccentric exercises as
described by Fiatarone et a1. (1990). Initial load was set at 50% of 1 RM for
3 sets of 10 repetitions, increasing to 80%. Resistance was provided by body
weight, sandbag weights and elastic bands. Maximum repetitions were
reassessed every 2 weeks and training adjusted to keep the load at 80%.
It appears that greater strength gains may be achieved when repetitions are
carried out without a rest between each repetition (Rooney et a1. 1994). Each
group of 10 repetitions comprises a set and the goal should be to repeat the set
three times with a short rest between each set. This dosage for strength train
ing applies whether the exercise utilizes free weights, elastic bands or practice
of sit-to-stand. Load is increased in STS by decreasing seat height by prescribed
increments. A person with very weak lower limb muscles can start with three
sets of a few repetitions from a higher than usual seat, increase number of repe
titions to 10, then progress to 10 repetitions from a lower seat.
The relationship between number of repetitions and amount of resistance is dif
ferent when the goal is endurance. In this case, high repetitions are practised at
low levels of resistance. For individuals following stroke, improving endurance
is important for a return to the community, so part of the exercise session
should involve this type of exercise, and can include treadmill walking and sta
tionary cycling. For the latter, the hand or foot can be bandaged to the handle
bar and pedal if necessary.
The individual's exercise programme should include vanatLOn in methods as
there is some evidence in the able-bodied that variation of stimulus has more
effect than repetition of the same movements under the same conditions
(Hakkinen and Komi 1981). Variation can involve a group circuit class, with
patients moving between work stations set up for specific types of exercise: step
ups, heels raise and lower, stationary cycling, treadmill walking with or without
harness, free weights, elastic band exercises, exercises on an isokinetic dynamom
eter, and leg and arm press machines. Pneumatic strength training equipment such
as the Keiser':' enables concentric and eccentric exercise including the leg press,
"Keiser Sports Health Equipment Inc., Fresno, CA.
Box 7.1 Summary of strength training
Strength training is carried out with sub-maximal loads - as a general rule
10 repetitions at 50-80% of maximal possible 1RM load, with a goal of 3 sets
Strength training utilizes resistance from body weight, free weights, elastic bands,
isokinetic dynamometry, exercise machines, treadmill walking
Exercise dosage is increased by increasing repetitions, number of sets, load
Strength training is task specific or oriented towards characteristics of tasks to
be learned
For endurance, high repetition numbers are practised at low levels of load, and
include stationary cycling, arm cycling and treadmill walking
For very weak muscles, the methods used are those which best facilitate force
generation for that individual and may include simple exercises, biofeedback,
mental practice, electrical stimulation
Strength training can be carried out under supervision, independently and in
group circuit training classes
with vanatIOns in posItIOn as required. The class is made enjoyable by the
company of others and the opportunity for socialization, two factors identified
as key components of successful programmes (Barry and Eathorne 1994).
In addition, exercise classes can provide increased levels of mental stimulation.
The class can be supervised by therapists and aides and is a time-efficient way
of organizing intensive training. Some patients will need close supervision,
others will not. Provision of harness support may alleviate fear of falling and
facilitate practice.
Adequate recovery time needs to be allowed so that muscles can adapt without
breaking down. A patient can, within a circuit training class, for example,
intersperse lower limb exercises with strengthening exercises for the upper limb
or practice of manipulation tasks, in this way exercising different muscles but
transferring energy and attention immediately from one action to another.
Patient participation and enthusiasm for exercise and training is encouraged by
the display of graphs of progress, videotapes of their motor performance and
photographs of performance to be achieved. Interest is sustained by ensuring
participants are always pushing toward their optimum performance, never
practising what is already easily achievable.
A summary of strength training is provided in Box 7.1.
Maximizing skill
Emphasis in trallllllg is on regallllllg optimal skill in particular functional
motor actions. Methods of maximizing skill in motor performance are based
on research into motor learning carried out with able-bodied subjects. In
skilled performance the goal is achieved successfully on most occasions, i.e. the
action is flexible enough to be performed under different environmental condi
tions, with minimal energy cost.
Following a brain lesion affecting the motor system, an immediate need is to
help the patient regain the ability to activate paretic muscles and generate force.
Strength training involves many repetitions under constrained conditions as a
means of increasing muscle contractility and strength and laying down a pattern
of coordination. Repetitive exercise may be as critical to motor learning follow
ing stroke as it is for healthy individuals learning to play the piano (Butefisch et
al. 1995). It is assumed that repetitive practice of a particular movement may
drive brain reorganization by what appears to be a process of motor learning
(Asanuma and Keller 1991). However, repetitive practice of the same move
ment may not be sufficient for promoting skill. While repetition is necessary for
increasing strength when muscles are weak, what Bernstein (1967) called 'repe
tition without repetition' is preferable when skill is to be acquired.
Strength training oriented toward the motor tasks in which competence must
be regained has the advantage of addressing both the reduced force generating
capacity and the regaining of skill. An example is the use of repetitive standing
up and sitting down (STS) as a strengthening exercise which also involves prac
tice of the action itself. Other examples illustrate the transfer effect of repeti
tive practice of an action that shares similar dynamic characteristics with other
actions. Repetitive practice of reaching to objects beyond arm's length in sit
ting was found to improve not only reaching performance but also STS (Dean
and Shepherd 1997). In this example, STS may have improved because the
patients repetitively practised reaching beyond arm's length, which involves
controlling upper body movement forward over the feet, a characteristic also
of the initial phase of standing up.
Methods of facilitating motor learning and skill training are described briefly
in Chapter 1 and in more detail elsewhere (Gentile 2000).
Measures of muscle strength
Dynamometry (grip force, hand-held, isokinetic dynamometers). A measurement of
peak isometric muscle force. In hand-held and isokinetic dynamometry the
subject does a maximum muscle contraction against a known force in a stand
ardized position (Mathiowetz et al. 1985, Bohannon et al. 1993, Murphy and
Spinks 2000). Isokinetic dynamometry can also be used to test multijoint and
more functional movement.
Other measures include:
Lateral Step Test (Worrell et al. 1993)
Medical Research Council (MRC) grades or Motricity Index (Medical
Research Council 1976, Wade 1992)
In able-bodied individuals, the length of muscles and other soft tissues is nor
mally sufficient for the activities regularly performed. Typically we spend some
time stretching (e.g. hip flexors, calf muscles) in order to increase our 'flexibil
ity' when we exercise or plan to engage in an unaccustomed sport. However,
flexibility declines with age (Roach and Miles 1991), partly due to changes in
collagen, a component of the fibrous connective tissue of ligaments and ten
dons. Length-associated changes can also occur after any prolonged period of
immobility, such as after an acute neural lesion. Stretching exercises are there
fore a critical part of a rehabilitation programme.
Increased range of motion after stretching may involve biomechanical, neuro
physiological and molecular mechanisms. Certain basic biomechanical proper
ties influence the muscle-tendon unit's response to stretch (Taylor et al. 1990).
Tension in muscle comprises passive and active components. Stretch can affect
the active component by altering neural activity and the passive mechanical
component by affecting viscous and elastic properties of the muscle. The mech
anisms of increasing muscle length and range of motion are not well known,
but may be found eventually in the cellular and adaptive mechanisms of a
muscle fibre (De Deyne 2001).
One characteristic of viscoelasticity is that the tissues respond to stretch and to
being held at a constant length with a decrease in tension, i.e., the stress or force
produced by the tissues at that length gradually declines, called 'stress relax
ation' (Duong et al. 2001). During stretch, soft tissues undergo progressive
deformation and they can be progressively extended with a constant force
which is called 'creep'.
Application of a passive stretching force affects an individual muscle according
to its anatomy, individual fibres receiving various amounts of stress depending
on their orientation relative to the longitudinal axis of the muscle (De Deyne
2001). Short-term increase in range of motion following a brief passive stretch
is explained by the viscoelastic behaviour of muscles and short-term changes
in muscle extensibility (Best et al. 1994). Longer term adaptive changes
(Halbertsma et al. 1996) that occur after sustained passive stretch produce a
longer muscle. The biological and molecular mechanisms by which this occurs
are not well understood; however, increase in sarcomere number has been
reported from animal experiments (Williams 1990).
Both decreased muscle length (contracture) and increased muscle stiffness are
common sequelae of stroke, appear to develop quickly and have potent negative
effects on the person's ability to exercise, train and regain effective performance
of motor actions. Contributors to loss of extensibility are disuse, degenerative
processes associated with ageing, and the effect of the stroke on the motor sys
tem. Loss of extensibility of calf muscles in particular can potentially have nega
tive effects on the ability to balance in standing, on walking, particularly up and
down stairs, and on standing up (Vandervoort 1999), and is said to predispose
the elderly to falls (Studenski et al. 1991).
Since contracture alters the resting limb position, it affects the resting length of
muscles crossing the joint and the lever torque angle relationship at the joint
(Lieber 1992). This alters the muscle's force output and can lead to muscle
imbalance (Vandervoort 1999). In some elderly subjects, joint limitation and
stiffness in the shoulders and spine, and some calf muscle shortening may have
been present before stroke.
Patients need to be monitored with diligence from immediately after a stroke,
since prevention of increased soft tissue stiffness and contracture is critical if
the negative effects are to be prevented. Intervention must be proactive and a
matter of routine within the hospital and rehabilitation centre. Patients who
are able to do unsupervised stretching and exercise should be advised to stretch
for a few minutes several times a day, particularly before and during training
sessions, and when they are practising on their own. Practice of simple exer
cises with elastic bands is one method of actively stretching muscles. Practice
of STS with the feet placed well back provides a simple method of actively
stretching the soleus muscle, provided weight is borne through that foot.
For patients who cannot voluntarily move a limb, passive means such as pos
itioning a limb or a joint are necessary. It has been shown that connective tissue
accumulation, a major contributor to increased muscle stiffness in shortened
muscles, can be prevented by passive stretching when muscles are inactive. For
example, short-term stretch (30 min daily) in animals prevented serial sarcom
ere loss, muscle atrophy and reduced range of motion (Williams 1990). In
humans, brief stretch of calf muscles can produce a decrease in stiffness
(Vattanasilp et al. 2000) and increased muscle compliance (Otis et al. 1985),
and is thought to 'precondition' a muscle prior to training (Taylor et al. 1990).
Although passive stretching may change the length of the muscle, it is active
stretching during dynamic task-oriented practice that affects the dynamic range
of a movement. One of the advantages of treadmill training is likely to be the
dynamic and task-specific stretch imposed on muscles in the stance limb such
as soleus, gastrocnemius and rectus femoris.
There is some evidence that passive stretching affects the active (neural) com
ponent of muscle tension, with a reduction of motoneuron excitability (Odeen
1981, Kukulka et al. 1986, Hummelsheim and Mauritz 1993). It is unlikely,
however, that decreasing reflex hyperexcitability by passive stretching has any
more than a short-term effect on reflex status, nor that it will be followed by
any improvement in motor control or function, if it is not also accompanied by
active stretching as part of functional training.
It is likely that the early institution of active stretch (through exercise and task
practice) may be a major factor in preventing the development of the adaptive
reflex hyperactivity associated with stiff, short and inactive muscles. There is
evidence that repetitive exercise and strength training can increase range of
motion and decrease resistance to passive movement (Blanpied and Smidt 1993,
Butefisch et al. 1995, Miller and Light 1997, Teixeira-Salmela et al. 1999).
Nevertheless, in patients with minimal active movement, muscles that can be
predicted to shorten and stiffen due to persistent maintenance of their resting
posture (e.g. calf muscles, long finger flexors, shoulder internal rotators and
adductors) need to be positioned for short periods of the day to preserve
length. Where contracture is present, progressive stretching by serial casting is
necessary (Moseley 1997). Injections of botulinum toxin may also be effective
Box 7.2 Summary of stretching
Prolonged passive stretching (15-30 min) via positioning during the day if a limb
cannot be actively moved, to prevent predictable muscle shortening and stiffness
(plantarflexors, shoulder adductors, internal rotators, elbow flexors, forearm
pronators, thumb adductors, long finger flexors). Daily passive stretch
(15-30 min) prevents atrophy and loss of sarcomeres in animals (Williams 1990)
Short (20 s) stretch to a stiff muscle(s), done manually by patient or therapist,
just prior to and during exercise. This can have the effect of preconditioning
the muscle(s) through stress relaxation and decreasing stiffness (Vattanasilp
et al. 2000)
Prolonged stretch to contracted soft tissues using serial casting (Moseley 1997),
combined with exercise and training carried out while casting is in place, with
follow-up exercise and training
in decreasing stiffness where it is extreme and accompanied by reflex hyper
activity. These methods must be combined with active exercise if the effects are
to be associated with functional improvement. So far there has been little
research into methods of prevention in stroke.
Stretching prescription
There is little basic information to guide the rational use of passive stretching
(Taylor et al. 1990). Although the optimal dosage for stretching is not clear,
there is evidence to suggest that holding a stretch for approximately 20 sand
repeating the stretch 4-5 times can be effective at lengthening the muscle-tendon
unit by inducing some stress relaxation and a reduction in the individual's per
ceived stiffness (Taylor et al. 1990). Prolonged stretch of 30 min has been
reported to decrease passive restraint in ankle plantarflexors in individuals with
paraplegia (Harvey and Herbert 2002). Passive joint mobilization may be neces
sary for individuals with spinal, glenohumeral or wrist joint pain and/or stiffness.
Stretching to preserve muscle length and flexibility is summarized in Box 7.2.
It should be noted that, although commonly used in some clinics, there is little
experimental support for the use of splinting (e.g. hand splints) to prevent or
correct length changes. The negative effects in terms of promoting learned non
use and altering the biomechanics of actions may outweigh any possible posi
tive effects. A splint worn at night may, however, be useful for those who do
not regain active use of the hand.
Measures of joint range
Dynamometer and goniometer. A known torque is applied to produce movement
in a standardized testing position. Range of movement is calculated from skin
surface markers. The use of polaroid photography, a calibrated spring-loaded
device and a goniometer have been found to be reliable for testing ankle range
of motion (Moseley and Adams 1991, Moseley 1997) (Fig. 7.4).
Strength training and physical conditioning 253
FIGURE 7.4 A torque controlled measurement procedure. Ankle angle is measured using skin
surface markers and photography. (Reproduced from Moseley 1997, with permission.)
Many individuals post-stroke are left with residual neurological deficits and
physiological changes in muscle fibres and muscle metabolism that are likely to
continue to impair function and impose excessive energy demands during the
performance of common daily activities, particularly walking. Metabolic
changes include decreased muscle blood flow, increased lactate production,
increased utilization of muscle glycogen and decreased capacity to oxidize free
fatty acids (Landin et a!. 1977). In addition, muscle fibre changes leading to a
decrease in oxidative metabolism and low exercise endurance can include a
decrease in proportion of type I muscle fibres due to the transformation of high
oxidative (type I) fibres to low oxidative (type II) fibres.
It is well documented that cardiovascular limitations lower exercise tolerance
to work capacity. A sedentary life-style leads to a further decline in muscle
strength and cardiovascular fitness. Complaints of fatigue attributed to disease
process are equally likely to be due to the demonstrably low levels of aerobic fit
ness and endurance. Several studies have reported that stroke subjects have less
energy, and experience increased social isolation and emotional distress when
compared with individuals of a similar age (Ahlsio et a!. 1984, Ebrahim et a1.
1986). However, until recently, there have been relatively few studies examining
the exercise (aerobic) capacity of those affected by stroke. Individuals post
stroke do not have the same functional capacity as healthy individuals due to
loss of physiological muscle mass. It is therefore difficult to use normative data
from healthy populations to predict capacity and guide training programmes
(Potempa et a1. 1996).
In a discussion of the rationale for aerobic exercise after stroke, Potempa and
colleagues (1996) summarize cardiovascular responses to exercise testing in
individuals after stroke, taken from available studies in chronic stroke. They
lowered peak exercise responses
reduced oxidative capacity of paretic muscles
decreased endurance.
Reduced functional capacity after stroke is therefore likely to be due, to varying
extents, to reduction in the number of motor units recruitable during dynamic
exercise, reduced oxidative capacity of weak muscles and low endurance, com
pounded in some individuals by the presence of comorbid coronary artery dis
ease and physical inactivity.
In spite of evidence that after stroke individuals are physically deconditioned,
aerobic exercise is not routinely prescribed for stroke patients, either early in
rehabilitation or after discharge (Smith et al. 1999). One of the reasons for this
may be that therapists have restricted any activities involving intensive effort, in
part because of a belief that effort increases spasticity, in part because many
patients are elderly. Neither point appears valid according to the evidence.
Effort applied in exercise does not increase spasticity or muscle stiffness. Elderly
individuals are capable of increasing their cardiovascular fitness, and improving
lifestyle and self-efficacy, with moderately vigorous exercise (Hamdorf et al.
1992, Mazzeo et al. 1998). In addition, aerobic exercise has the potential to
minimize secondary effects on muscle fibre transformation by enhancing motor
unit recruitment and favouring development of high oxidative fibres (Potempa
et al. 1996). Appropriate exercise may increase endurance capacity and min
imize symptoms of cardiovascular disease (Hamm and Leon 1994). Exercise
and physical activity are known to have positive psychological effects in the
able-bodied young and elderly, and in disabled individuals. There also appears
to be a relationship between physical fitness and cognition (Rogers et al. 1990,
Spirduso and Asplund 1995).
Recent studies have shown the benefits of physical conditioning programmes
following stroke, including improvements in both aerobic capacity and func
tional abilities (Potempa et al. 1995, 1996, Macko et al. 1997, Teixeira-Salmela
et al. 1999). There are reports of improved aerobic capacity with appropriate
training such as bicycle ergometry (Potempa et al. 1995, Bateman et al. 2001),
with graded treadmill walking (Macko et al. 1997) and with a combination of
aerobic and strengthening exercises (Teixeira-Salmela et al. 1999). Supervised
aerobic training programmes have been shown to improve V0
, with the
improvement significantly related to improvements in motor function (Potempa
et al. 1995). The evidence so far suggests that aerobic exercise on a bicycle
ergometer, treadmill walking or with graded walking significantly improves
physical fitness when individuals are tested on the training exercise (Table 7.1).
As might be expected, the effects are exercise specific. Generalization occurs,
however, in the improvements noted in general health and well-being. Any gen
eralization to performance of other actions probably depends on the patient's
level of impairment (Bateman et al. 2001).
TABLE 7.1 Conditioning: clinical outcome studies
Reference Subjects Methods Duration Results
Potempa et al. 42 Ss Randomized controlled 10 weeks Only E showed significant
1995 >7 months trial 3 days/week, improvements on maximum
post-stroke E: adapted cycle 30min oxygen consumption, workload
Age range ergometer (training load and exercise time, lower
21-77 yrs was gradually increased systolic blood pressure at
from workload at 30-50% submaximal workloads during
of maximum effort to the graded exercise test. Aerobic
highest level attained by capacity was related to
the subjects) improvement in function
C: passive ROM (Fugl-Meyer Scale)
Macko et al. 9 Ss Pre-test, post-test design 6 months Significant reduction in energy
1997 >6 months Low intensity aerobic 3 days/week, expenditure (p < 0.02),
post-stroke exercise (graded treadmill 40min exercise-mediated decline in
Mean age training at 50-60% heart oxygen consumption (p < 0.02)
67 2.8 yrs rate reserve; warm up, and respiratory exchange ratio
cool down period) (p < 0.01) during standardized
1mph submaximal effort
treadmill walking
Smith et al. 1998 14 Ss Pre-test. post-test design 3 months Significant increases in
>6 months Low intensity aerobic 3 days/week, quadriceps torque (p < 0.01)
Mean age exercise (graded treadmill 40min generated across all angular
66 3 yrs training at 50-60% heart velocities in eccentric and
rate reserve; warm up, concentric modes bilaterally.
cool down period) Affected limb: eccentric mode
Resisted or strengthening 51%; concentric mode 38%
exercises on isokinetic Non-affected limb: eccentric
dynamometer (at 4 angular mode 25%; concentric
velocities 30, 60, 90, mode 17%
Silver et al. 2000 5 Ss Pre-test. post-test design 3 months Significant reduction in time
>6 months Aerobic exercise (graded 3 days/week, taken to perform modified Get
post-stroke treadmill walking at 40min Up and Go Test (p < 0.05).
Mean age approximately 60-70% increased gait velocity (p < 0.01)
60.4 2.7 yrs heart rate reserve) and cadence (p < 0.05). No
change in gait symmetry during
Get Up and Go Test
C. control group; E. experimental group; ROM, range of movement; 55, subjects.
Since aerobic training has the potential to enhance the development of high
oxidative muscle fibres and motor unit recruitment, it is interesting that there is
promising evidence that strength training and physical conditioning programmes
result in significant improvements in performance of functional activities. For
example, Silver and colleagues (2000) reported that an aerobic treadmill training
programme, three times a week for 3 months, resulted in an increase in gait
velocity and cadence, and a decrease in the overall time required to perform a
modified 'Get Up and Go' test. Exercise intensity was individualized and
advanced to 40 minutes' training at approximately 60-70% of maximum heart
rate reserve. Teixeira-Salmela et al. (1999) assessed subjects' general level of physi
cal activity on the Human Activity Profile, a survey of 94 activities including trans
portation, home maintenance, social and physical activities, which are rated
according to their required metabolic equivalents. The results indicated that sub
jects were more able to perform household chores and recreational activities after
strengthening and aerobic training. These subjects also reported an improved qual
ity of life, as previously reported some years ago (Brinkman and Hoskins 1979).
Although clinical evidence of cardiovascular disease may limit exercise and
training initially, Potempa and colleagues (1996) suggest that a monitored aer
obic training programme may improve endurance and functional ability and
have the following physiological benefits:
increased work capacity
decreased resting and sub-maximal heart rates and blood pressure
weight loss
improved lipoprotein profile
decreased platelet aggregation
delay in onset of angina.
A recent study has shown the early deconditioning that occurs within the first
6 weeks after stroke (Kelly et al. 2002) and illustrates the need for exercise
of greater intensity than is typically provided. The study also shows that,
for the patients investigated, exercise testing was well tolerated at this early
stage. Patients performed incremental maximal effort tests on semi-recumbent
cycle ergometers (see Fig. 3.26), with continual monitoring of heart rate and
blood pressure by a physician.
Exercise prescription
Once intensive exercise can begin, patients can work on a cycle ergometer, with
adaptive devices as necessary (e.g. foot can be strapped to pedal) (Potempa et al.
1995), on a device such as the Motomed Leg Trainer *, or on a treadmill. As an
example of exercise prescription suitable for this patient group, Potempa et al.
(1996) adapted the criteria of the American College of Sports Medicine for
patients with low functional capacity (Kenney et al. 1995) for their conditioning
Initially patients train at a workload equivalent of 40-60% of V0
gressing up to 30 min, 3 times/week.
When 30 min is reached, intensity is progressively increased to the highest
workload tolerable without symptoms.
10 min warm-up and cool-down periods of unloaded cycling.
Another study describes a similar programme of graded treadmill walking
(Macko et al. 1997).
Changes in muscle function, particularly in the elderly, are not maintained with
out ongoing training (Fiatarone et al. 1990). Ongoing post-discharge exercise
*MOTOmed PieD Leg Trainer, Reck, Reckstrasse 1-3, D-88422, Betzenweiller, Germany.
programmes, either home based, in which case they need monitoring at regular
intervals, or gymnasium or health centre based, need to be available for indi
viduals following stroke, particularly when they are elderly. Commenting on
programmes for the healthy elderly, Mazzeo and colleagues (1998) point out that
exercise programmes for strength and fitness should consist of scientifically
based strategies rather than the non-specific 'movement' programmes typically
As a general health precaution it is advisable for older individuals to have a
medical check prior to starting a moderately vigorous exercise and fitness pro
gramme. This should also occur after stroke. The contraindications to exercise
training appear to be few, even in the very old (Mazzeo et al. 1998). Certain
conditions such as febrile illness or unstable chest pain need investigation.
Patients with arthritis can also tolerate strength training well (Lyngberg et al.
1988). Exacerbation of joint pain from underlying arthritis may necessitate
modification of exercise; for example, closed-chain exercise may be preferable
to open-chain exercise for those with infrapatellar pain.
Measures of exercise response
Maximal effort exercise testing
Standardized maximal treadmill or cycle ergometer test After stroke, exercise test
ing is typically carried out using a treadmill or bicycle ergometer. A stationary
cycle is considered by some to be preferable to treadmill or arm-mobilized
ergometer for testing after stroke (Potempa et al. 1995). A cycle ergometer
makes it easier to quantify external workload. A standardized methodology,
such as the Balke Protocol (Fletcher and Schlant 1994), is used (Macko et al.
1997). Maximal testing requires simultaneous electrocardiograph monitoring
and is carried out by specially trained staff. Tests of V0
peak and O
sumption per kilogram body weight at a given power output while cycling,
graded walking or treadmill walking provide a means of evaluating cardiovas
cular response to exercise (Jankowski and Sullivan 1990).
Submaximal effort exercise testing
Standardized submaximal treadmill test. Macko and colleagues (1997) describe a
treadmill test (1 mph, no incline) representative of the slow walking typical
after stroke, performed with open circuit spirometry to measure oxygen con
sumption (V0
) under steady state conditions. Patients wore a nose clip and
breathed through a mouthpiece, or were fitted with a non-rebreathing mask to
collect expired air. All procedures must be standardized.
Monitoring exercise level during training
Standardized heart rate test A simple measure of heart rate (HR) gives an indica
tion of intensity of exercise. It can be used to monitor level of exercise to ensure
it is sufficiently vigorous and as a simple test of whether or not the patient's
cardiovascular system is adapting to exercise. The target HR is determined using
a HR monitor:
Calculate maximum age-predicted HR by subtracting the patient's age
from 220
Maximum age-predicted HR = 220 - age
Calculate target HR as between 60 and 80% of maximum
age-predicted HR
Target HR range = 60-80% X maximum age-predicted HR
Endurance can be tested specifically for walking by the 6-minute or 12-minute
walk tests (Guyatt et al. 1985), in which the distance walked in 6 or 12 min
utes is measured.
Participation in exercise programmes designed to increase strength and fitness
and preserve soft tissue length and flexibility can result in significant improve
ments in strength, functional motor performance and physical fitness in indi
viduals following stroke. There may be additional benefits in terms of attitude,
self-concept, self-efficacy, cognition, and confidence in engaging in physical
actions, together with recognition of functional potential and the ability to
deal with disabilities more effectively (Mazzeo et al. 1998, Teixeira-Salmela
et al. 1999). The regaining of skill in critical tasks requires specific training, with
intensive practice of actions in the appropriate contexts. The individual must,
however, be fit enough to perform the tasks of daily life.
Mazzeo et al. (1998) in their position paper on exercise in the elderly comment
that 'sedentariness' appears a far more dangerous condition than physical
activity in the elderly. They point out that a large amount of data dispels
myths of futility of exercise and provides reassurance about the safety of exer
cise in the oldest adults. Following discharge from rehabilitation, the potential
for negative effects from immobility and inactivity following stroke should be
an ongoing public health issue.
One focus of future research needs to be on the retention of gains after the
completion of exercise programmes. It is likely that older individuals with
stroke need to continue with both home-based and periodic supervised group
exercise in order to retain functional gains and reduce the functional declines
associated with ageing. Exercise for this group, as is advocated for the healthy
elderly, appears critical to offset loss of muscle mass, improve bone density,
decrease likelihood of falls and reduce the risk factors associated with, for
example, diabetes and cardiovascular disease.
If individuals are to make their best possible recovery from the effects of
stroke, they should initially be in a stroke unit with specially trained staff. In
this environment they can receive appropriate diagnostic and medical interven
tions to reduce the likelihood of complications. Stroke unit care has been
shown to be associated with a reduction in death and dependency which was
independent of such factors as patient age and sex (lndredavik et al. 1997,
Stroke Unit Trialists' Collaboration 1997). A stroke unit provides opportuni
ties for early active intervention. A specialized unit can offer support, encour
agement and education programmes for patients and their families with staff
trained to coordinate multidisciplinary rehabilitation.
Immediately after stroke it is difficult to predict the extent of eventual recovery.
After about 2 weeks post-stroke, good prognostic signs include urinary con
tinence, younger age, mild stroke, rapid improvement, good perceptual abilities
and no cognitive disorders. Significant predictors of poor outcome appear to be
persistent incontinence and poor premorbid functioning (Cifu and Stewart
1996). However, there are patients who improve unexpectedly and others who
do poorly despite having a good prognosis.
Early and proactive physiotherapy can reduce the likelihood of negative sequelae
such as soft tissue contracture, learned non-use and persistence and habituation
of perceptual--cognitive impairments. Active training in sitting in the acute phase
is critical to prevent complications associated with the supine position and bed
rest. It is well established that early mobilization can itself reduce secondary
thromboembolic events, pneumonia and mortality in stroke. Getting the person
upright and active is also necessary to activate the patient's attentional state. It is
possible that the longer the delay in starting active training of balancing in sit
ting, the more likely it is that the patient will become fearful and apprehensive in
any attempt at dealing with gravity. Even a very weak patient can be assisted to
turn to one side, to sit up over the side of the bed, and to actively practise bal
ancing in sitting (Fig. 8.1 ). Rehabilitation starts, therefore, in the acute phase.
The rate and extent of mobilization depend on the patient's condition. However,
active training in sitting and standing is commenced as soon as vital signs are
260 Appendices
FIGURE 8.1 Lateral neck flexion is encouraged as he is assisted to sit over the side of the
bed. Ensure that there is a firm surface for him to put his feet on if they do not
reach the floor.
stable (Fig. 8.2). If medical complications preclude sitting out of bed and active
mohilization, regular and frequent turning is required to maintain skin integrity,
prevent aspiration and pooling of secretions in the lungs. While the patient is
confined to bed, positioning of the limbs with repetitive passive and active
range of motion exercises are instituted with the aim of preventing stiffness.
Care should be taken when passively moving the shoulder. Any movement into
elevation should include external rotation.
Patients and their families are given information and provided with the oppor
tunity to [earn about the stroke, the process of rehabilitation and their role in
this. Since the patient and family will be distressed and anxious in the early
stages, the opportunity to gain information needs to be ongoing. Members of
stroke support groups, who have themselves experienced a stroke, can be help
ful in talking with the patient.
The impact of stroke on many body systems is particularly evident in the very
early stages. Since rehabilitation is carried out by a multidisciplinary team, it is
necessary for all members to understand these complications and their overall
management. Five common complications are discussed briefly below. The reader
should refer for more detailed information about the management of these
and other complications associated with stroke (Gresham et al. 1995, Carr and
Shepherd 1998).
Overview 261
FIGURE 8.2 In her room, both bed and chair are adjusted to a suitable height to enable practice
of standing up and sitting down. (a) Therapist stabilizes the foot. (b) Stand-by
(a) (b)
Dysphagia, or difficulty swallowing, is a relatively common problem following
stroke and may lead to aspiration of saliva, food or liquids. The prevalence is
hard to determine. Early detection and treatment is important in preventing
both aspiration and dehydration. Primary deficits are incoordination of the
oral and pharyngeal phases of swallowing, and delayed or absent triggering of
the swallow reflex, or both. Specific training of orofacial musculature to assist
chewing and swallowing safely should be commenced in sitting as soon as dys
phagia is identified. A modified diet the consistency of mashed potato may aid
the re-establishment of control of swallowing. Nasogastric tube feeding may be
required where dysphagia is severe. Videofluoroscopy using a modified barium
swallow may be used to evaluate the pharyngeal phase of swallowing and the
mechanism of aspiration in patients whose dysphagia persists beyond the early
stage of recovery.
Urinary incontinence is common but usually transient following stroke. It can be
related to motor and sensory impairments, difficulty communicating (which
interferes with making the need to void known to others) and immobility, as well
as to cognitive deficits and inability to recognize the need to void. There is some
evidence of neurological deficits leading to retention or overflow or both (e.g.
Gelber et al. 1993) and a catheter may be required. Early assumption of standing
262 Appendices
and early ambulation usually help the individual to overcome incontinence. If it
does not, a bladder training programme can be instituted. Persistent incontinence
may be secondary to poor cognitive function (Wade et al. 1985) and suggests a
poor prognoSIs.
Communication deficits. The aphasias (usually dysphasia, since there is rarely a
total loss of language) are disturbances of language. They are distinct from motor
disorders of speech resulting from weakness and incoordination of the muscles
controlling the vocal apparatus, which is classified as dysarthria. Communication
deficits isolate the patient and may, not surprisingly, lead to anger and frustra
tion as attempts to communicate fail. It is critical to establish communication
with the patient through gesture, situational cues, visual prompts, facial expres
sion, short simple sentences and eye contact. Care should be taken not to exclude
the patient from the conversation. Individuals with language impairment usually
experience some degree of spontaneous improvement (Wade et al. 1986). A
speech and language professional identifies the type and severity of the problem
and can provide advice on the best methods of communication.
FIGURE 8.3 Two slide sheets (Hemco MFG Industries, Ballarat, Australia) are used to move the
patient up in the bed so she will be sitting upright when the top of the bed is elevated.
Overview 263
Injury to the paretic, unprotected shoulder can occur through lifting the patient
under the arms or pulling on the paretic arm. These lifting methods should be
replaced by use of slide sheets to move the patient up in the bed (Fig. 8.3), a
hoist when sitting a very weak patient out into a chair, and a slide board
(Patboard) to move a patient from hospital bed to exercise table (Fig. 8.4a).
Positioning the shoulder is particularly important when the patient is sitting in a
chair (see Fig. 5.23).
Excessive attentional focus to the right side occurs in patients with visuospatial
agnosia (neglect). This disorder is complex (eh. 7). The head and eyes may be
strongly deviated to the right with an apparent inability to turn the head and
eyes to the left. It appears that the person pays attention only to right-sided cues.
The patient's deficit is brought to their attention in a way that motivates
attempts to offset the deficits, rather than covering up for what they cannot
understand. Strong, consistent cues are required to orient the patient to turn the
head and eyes and pay attention to the left. Features of a task and the relevant
FIGURE 8.4 (a) Slide board and sheets are used to move an acute patient on to a firm height
adjustable surface to practise sitting balance. (b) Sitting practice: therapist prevents
his L foot from sliding forward. He could only hold his head up momentarily - a soft
collar would provide a more erect head position to facilitate spatial orientation and
eye contact with the person in front of him. Here he is trying to move the upper body
forward by flexing at the hips and sliding hands forward on the table. During previous
attempts to sit on the more compliant surface of his bed, two people were needed to
prevent him from falling. Here, the firmer surface and the table enable him to be
trained by one person.
(a) (b)
cues which must be attended to for successful completion are highlighted. The
patient is asked to search for a particular object (concrete goal) rather than being
repetitively reminded to turn the head to the left (abstract goal). A close family
member who understands the deficit may be able to assist the person to re-orient
the head and eyes leftwards. Extraneous stimuli on the right side such as an unat
tended television or radio should be avoided. Excessive attentional focus may
improve quite quickly. Persistent visual inattention complicates rehabilitation.
As soon as possible, the focus changes from a medical and sickness orientation
to emphasize exercise and functional training planned around the patient's
greatest need, which is to regain effective functioning in daily life. Rehabilitation
is specific for each individual's needs, and includes not only a motor training
programme, but also programmes designed to overcome specific visual, cogni
tive, perceptual, swallowing, communication and continence problems. The pre
vention of complications of immobility and disuse, including reduced endurance,
cardiovascular fitness, and the facilitation of psychosocial adjustment to disabil
ity and community integration, are central to rehabilitation management of indi
viduals with stroke.
Rehabilitation and the environment in which it occurs is planned to take account
of the patient as active learner. The patient's day should provide opportunities to
regain self-confidence, take responsibility for unsupervised practice, and regain a
sense of time. An overall plan is put in place to motivate the patient and stimu
late the individual's learning abilities in both motor and cognitive functions. The
therapist's ability to establish contact with the patient through dialogue and posi
tive feedback helps to generate a positive atmosphere and increase the patient's
confidence (Talvitie 2000). Furniture, equipment and activities are designed to
enable active practice of the necessary tasks outside the protective atmosphere of
the rehabilitation unit. Delivery of rehabilitation services is planned to maximize
time spent in preparation for the post-discharge phase.
Staff need to be aware that the typical rehabilitation environment is complex,
unfamiliar and unpredictable to those who suddenly find themselves in it. Patients
have to get to know and interact with a number of different health professionals
and other patients. This can be a stressful situation in itself, but after suffering a
stroke, a person with a severe disability also has to cope with their own sense of
loss and anxiety about the future. For a previously self-sufficient person, adapting
both to the disability and to the rehabilitation environment may be difficult and
can erode the feeling of competence.
The general role of physiotherapy is to help the person to become physically
and mentally active again, with a flexible and strong musculoskeletal system. It
is the physiotherapist's role to encourage active participation in the rehabilita
tion process and to assist the patient to regain the pleasures inherent in being
physically and mentally active.
The aims of physiotherapy following stroke are to optimize:
functional motor performance
physical fitness, strength and endurance
interest and motivation
mental and physical vigour
and to prevent:
secondary neuromusculoskeletal and cardiovascular disability related to
physical inactivity and disuse.
Given our contemporary understanding of brain reorganization and what drives
it, plus the emerging empirical evidence of what methods of intervention work
and what do not, our recommendations for physiotherapy include the following.
Start active task-oriented exercise and training of everyday actions (sit
ting balance, reaching and manipulation, sit-to-stand, standing balance
and walking) as soon as vital signs are stable, with patient dressed in
clothes suitable for exercising.
Outline the process of rehabilitation and discuss the patient's and family's
roles in this. Explain the necessity for intensive and repetitive practice.
Institute stretching protocols for at-risk muscles, in particular soleus,
shoulder internal rotators and adductors, forearm pronators, wrist and fin
ger flexors, thumb flexors and adductors.
Estimate strength of key muscle groups and start repetitive resisted
strength training, for example, lower limb extensors, shoulder flexors,
abductors and external rotators, wrist and finger extensors. Use exercise
equipment as required.
Institute electrical stimulation and EMG feedback where necessary.
Start practice book and check list of what patient is to practise in order to
establish a daily exercise routine.
Provide seating suitable for individual's body dimensions and needs, for
example, height-adjustable seat for exercising, suitable chair for sitting,
with arm rest (glenohumeral joint in mid-rotation) (see Fig. 5.23).
Test and record baseline motor performance on everyday actions; repeat
at intervals, on discharge and follow-up.
The emphasis in rehabilitation research is typically on inpatient care. However,
wider issues need to be addressed, especially as long-term disability and social
and emotional sequelae of stroke affect the quality of life after discharge.
Follow-up services may be poor or non-existent. Many individuals are reported
to suffer a loss of rehabilitation gains after they are discharged from rehabilita
tion (Paolucci et al. 2001). A recent survey in the UK showed a paucity of pro
vision of follow-up services beyond the most basic level of community therapy
or support. This was compounded by poor communication between profession
als and between patients, carers and professionals, high levels of patient dissat
isfaction, and low expectations by professionals of patients' abilities (Tyson and
Turner 2000). The significance of this study lies in its methodology which was
able to identify not only the shortcomings of services provided but also the rea
sons for service failure.
Major oversights in post-discharge stroke management are poor access to trans
port and lack of community exercise facilities for the disabled. There has been
little emphasis on the obvious need for individuals with disability to continue
with post-rehabilitation exercise and training to not only maintain but also to
progress their functional abilities and levels of physical fitness. We know from
recent studies of individuals post-discharge that many people, even several years
after stroke, are able to make positive gains in strength, fitness, endurance and
skill. It has been suggested that the stress of adjusting to the major life change
caused by stroke itself may interfere with full potential being achieved during
initial inpatient rehabilitation. Patients may have increased motivation to
improve function once in their home environment, where the impact of limited
functional ability becomes more obvious (Tangeman et al. 1990). Without
ongoing or 'top-up' exercise programmes to maintain and improve functional
ability, however, rehabilitation gains can be lost once the individual is outside
the structure of the rehabilitation setting.
Disability has serious consequences beyond physical function. Difficulty per
forming essential everyday actions can impose continuing physical inactivity
and social isolation on disabled individuals, affecting well-being and quality of
life. Psychological sequelae may include loss of self-esteem, fear of walking
outside the house, or fear of falling. Social consequences may include role
losses, isolation and increased demands on caregivers.
As leisure, sport and social activities are important to quality of life they need to
be included as an integral part of planning for discharge and to continue
beyond discharge. A community-based stroke group can play an important role;
for example, the Stroke Recovery Association of Australia is a social and self
help organization for stroke persons and their families. Weekly group meetings
organized by members at different locations offer a rewarding experience for
people who have had a stroke, providing emotional support and assisting in the
transition back into the community, and in some cases offering a weekly exer
cise group.
Recent evidence of the effectiveness of task-oriented trammg and muscle
strengthening is promising. It is very likely that an increased understanding of
human movement and of the effects of training in a stimulating and challenging
environment will begin to give a clearer picture of the potential for brain reorgan
ization and functional recovery after an acute brain lesion.
Abercrombie MLJ, Lindon RL, Tyson MC (1964)
Associated movements in normal and physically
handicapped children. Dev Med Child Neurol,
Ada L, O'Dwyer N (2001) Do associated reactions
in the upper limb after stroke contribute to
contracture formation? Clin Rehabil, 15,
Ada L, Westwood P (1992) A kinematic analysis of
recovery of the ability to stand up following
stroke. Aust I Physiother, 38, 135-142.
Ada L, O'Dwyer NJ, Neilson PD (1993)
Improvement in kinematic characteristics and
coordination following stroke quantified by
linear systems analysis. Human Mov Sci, 12,
Ada L, Vattanasilp V, O'Dwyer NJ et al. (1998)
Does spasticity contribute to walking
dysfunction after stroke? I Neural Neurosurg
Psychiatry, 64, 628-635.
Ada L, Mackey F, Heard Ret al. (1999) Stroke
rehabilitation: does the therapy area provide
a physical challenge? Aust I Physiother, 45,
Ada L, Canning C, Dwyer T (2000) Effect of
muscle length on strength and dexterity after
stroke. Clin Rehabil, 14,55-61.
Adams JM, Perry J (1994) Gait analysis: clinical
application. In Human Walking (eds J Rose, JG
Gamble), Williams & Wilkins, Philadelphia,
2nd edn, pp 139-164.
Adams R (1990) Attention control training and
behaviour management. In Key Issues in
Neurological Physiotherapy (eds L Ada, C
Canning), Butterworth-Heinemann, Oxford,
pp 81-97.
Adams RW, Gandevia SC, Skuse NF (1990) The
distribution of muscle weakness in upper
motoneuron lesions affecting the lower limb.
Brain, 113, 1459-1476.
Agahari I, Shepherd RB, Westwood P (1996)
A comparative evaluation of lower limb forces
in two variations of the step exercise in able
bodied subjects. In Proceedings of 1st
Australasian Biomechanics Conference (eds
M Lee, W Gilleard, P Sinclair et al.), University
of Sydney, NSW, Australia, pp 94-95.
Ahlsio B, Britton M, Murray V et al. (1984)
Disablement and quality of life after stroke.
Stroke, 15, 886-890.
Akeson WH, Amie! D, Abel MF et al. (1987)
Effects of immobilization on joints. Clin Orthop
Relat Res, 219, 28-37.
Andrews AW, Bohannon RW (2000) Distribution
of muscle strength impairments following
stroke. Clin Rehabil, 14, 79-87.
Andrews AW, Thomas MW, Bohannon RW
(1996) Normative values for isometric
muscle force measurements obtained with
hand-held dynamometers. Phys Ther, 76,
Andriacchi TP, Ogle JA, Galante JO (1977)
Walking speed as a basis for normal and
abnormal gait measures. I Biomech, 10,
Andriacchi TP, Andersson GBJ, Fermier RW et al.
(1980) A study of lower-limb mechanics during
stair-climbing. I Bone It Surg, 62A, 749-757.
Anianson A, Gustafsson E (1981) Physical training
in elderly men with reference to quadriceps
muscle strength and morphology. Clin Physiol,
Anianson A, Grimby G, Rundgren A et al. (1980)
Physical training in old men. Age Ageing, 9,
AnnettJ (1971) Acquisition of skill. Br Med Bull,
Arbib MA, Iberall T, Lyons D (1985) Coordinated
control programs for control of the hand. In
Hand Function and the Neocortex (eds AW
Goodwin, I Darian-Smith), Experimental Brain
Research Suppl 10, Springer, Berlin,
pp 111-129.
Arsenault AB, Winter DA, Martenuik RG et al.
(1986) Treadmill versus walkway locomotion in
humans: an EMG study. Ergonomics, 29,
Arsenault AB, Bilodeau M, Dutil E et al. (1991)
Clinical significance of the V-shaped space in the
subluxed shoulder of hemiplegia. Stroke, 22,
Aruin AS, Latash ML (1996) Anticipatory postural
adjustments during self-initiated perturbations
of different magnitude triggered by a standard
motor action. Electroencephalagr Clin
Neurophysiol, 101,497-503.
Arutyunyan GA, Gurfinkel VS, Mirskii ML (1969)
Organization of movements on execution by
man of an exact postural task. Bio(izika, 14,
Asanuma H, Keller A (1991) Neuronal mechanisms
of motor learning in mammals. Neurareport, 2,
Ashburn A (1995) Behavioural deficits associated
with the "pusher" syndrome. Proceedings of
World Congress of Physiotherapists,
Washington, DC, p 819.
Ashworth B (1964) Preliminary trial of
carisoprodol in multiple sclerosis. Practitioner,
192, 540-542.
Asmussen E (1953) Positive and negative muscular
work. Acta Physiol Scand, 28, 364-382.
Asmussen E (1980) Aging and exercise.
In Environmental Physiology: Aging, Heat
and Altitude (ed J Horvath), Elsevier
North-Holland, Amsterdam, pp 419-428.
Asmussen E, Bonde-Petersen F (1974) Storage of
elastic energy in skeletal muscles in man. Acta
Physiol Scand, 91, 385-392.
Baker LL, Yeh C, Wilson D et al. (1979) Electrical
stimulation of wrist and fingers for hemiplegic
patients Phys Ther, 59, 495-499.
Balmaseda MT, Koozekanani SH, Fatehi MT
et al. (1988) Ground reaction forces, center of
pressure, and duration of stance with and
without an ankle-foot orthosis. Arch Phys Med
Rehabil, 69,1009-1012.
Barry H, Eathorne S (1994) Exercise and aging:
issues for the practitioner. Med Clin North Am,
Bassey EJ, Fiatarone MA, O'Neill EF et al. (1997)
Leg extensor power and functional performance
in very old men and women. Clin Sci, 82,
Batemen A, Culpan FJ, Pickering AD et al. (2001)
The effect of aerobic training on rehabilitation
outcomes after recent severe brain injury: a
randomized controlled trial. Arch Phys Med
Rehabil, 82, 174-182.
Becher JG, Harlaar J, Lankhorst GJ et al. (1998)
Measurement of impaired muscle function of
the gastrocnemius, soleus, and tibialis anterior
muscles in spastic hemiplegia: a preliminary
study.] Rehabil Res Dev, 35, 314-326.
Beckerman H, Becher J, Lankhorst GJ (1996)
Walking ability of stroke patients: efficacy of
tibial nerve blocking and a polypropylene
ankle-foot orthosis. Arch Phys Med Rehabil,
Beer RF, Given JD, Dewald JpA (1999) Task
dependent weakness at the elbow in patients
with hemiparesis. Arch Phys Med Rehabil, 80,
Belenkii VY, Gurfinkel VS, Pal'tsev YI (1967)
Elements of control of voluntary movements.
Bio{izika, 12, 135-141.
Bendz P (1993) The functional significance of the
fifth metacarpus and hypothenar in two useful
grips of the hand. Am] Phys Med Rehabil, 72,
Bennett EL (1976) Cerebral effects of differential
experience and training. In Neural Mechanisms
of Learning and Memory (eds MR Rosenzweig,
EL Bennett), MIT Press, Cambridge MA,
pp 297-287.
Berg K, Wood-Dauphinee S, Williams JI et al.
(1989) Measuring balance in the elderly:
preliminary development of an instrument.
Physiother Can, 41, 304-311.
Berg KO, Maki BE, Williams JI et al. (1992)
Clinical and laboratory measures of postural
balance in an elderly population. Arch Phys
Med Rehabil, 73, 1073-1080.
Berger W, Horstmann G, Dietz V (1984) Tension
development and muscle activation in the leg
during gait in spastic hemiparesis: independence
of muscle hypertonus and exaggerated stretch
reflexes.] Neural Neurosurg Psychiatry, 47,
Berger RA, Riley PO, Mann RW et al. (1988)
Total body dynamics in ascending stairs and
rising from a chair following total knee
arthroplasty. Proceedings of the 34th Annual
Meeting of the Orthopedic Research Society,
Atlanta, Georgia, p 542.
Bergner M, Bobbitt RA, Carter WB et al. (1981)
The sickness impact profile: development and
final revision of a health status measure. Med
Care, 19, 787-805.
Bernhardt J, Ellis P, Denisenko S (1998) Changes
in balance and locomotion measures during
rehabilitation following stroke. Physiother Res
lnt, 3,109-122.
Bernstein N (1967) The Coordination and
Regulation of Movements. Pergamon Press,
Best TM, McElhaney J, Garrett WE et al. (1994)
Characterization of the passive responses
of live skeletal muscle using the quasilinear
theory of viscoelasticity.] Biomech, 27,
Bisiach E, Vellor G (1988) Hemineglect in
humans. In Handbook of Neuropsychology
(eds F Bollar, J Grafman), Elsevier, Amsterdam,
pp 195-222.
Blanton S, Wolf S (1999) An application of
upper-extremity constraint-induced therapy in
a patient with subacute stroke. Phys Ther, 79,
Blanpied P, Smidt GL (1993) The difference in
stiffness of the active plantarflexors between
young and elderly human females. ] Gerontol,
48, M58-M63.
Blundell SW, Shepherd RB, Dean CM et al. (2002)
Functional strength training in cerebral palsy. A
pilot study of a group circuit training class for
children aged 4-8 years. Clin Rehabil (in press).
Bobath B (1990) Adult Hemiplegia: Evaluation and
Treatment, 3rd edn, Butterworth-Heinemann,
Bobbert MF, van Soest AJK (2000) Two-joint
muscles offer the solution, but what was the
problem? Motor Control, 4, 48-52.
Bogataj U, Gros N, Malezic M et al. (1989)
Restoration of gait during two to three weeks of
therapy with multichannel electrical stimulation.
Phys Ther, 69, 319-327.
Bohannon RW (1988) Muscle strength changes in
hemiparetic stroke patients during inpatient
rehabilitation. ] Neurol Rehabil, 2, 163-166.
Bohannon RW (1996) Letter: Ipsilateral pushing in
stroke. Arch Phys Med Rehabil, 77, 524.
Bohannon RW (1998) Correction of recalcitrant
lateropulsion through motor relearning. Phys
Ther Case Reports, 1, 157-159.
Bohannon RW (1999) Observations of balance
among elderly patients referred to physical
therapy in an acute care hospital. Physiother
Theory Pract, 15, 185-189.
Bohannon RW, Andrews AW (1990) Correlation
of knee extensor muscle torque and spasticity
with gait speed in patients with stroke. Arch
Phys Med Rehabil, 71, 330-333.
Bohannon RW, Smith MB (1987) Interrater
reliability of a modified Ashworth Scale of
muscle spasticity. Phys Ther, 67, 206-207.
Bohannon RW, Walsh S (1991) Association of
paretic lower extremity muscle strength and
standing balance with stair-climbing ability in
patients with stroke. ] Stroke Cerebrovasc Dis,
Bohannon RW, Larkin PA, Smith MB et al. (1986)
Shoulder pain in hemiplegia: statistical
relationship with five muscles. Arch Phys Med
Rehabil, 67, 514-516.
Bohannon RW, Walsh S, Joseph MC (1993)
Ordinal and timed balance measurements:
reliability and validity in patients with stroke.
Clin Rehabil, 7, 9-13.
Bohannon RW, Smith J, Hull D et al. (1995)
Deficits in lower extremity muscle and gait
performance among renal transplant candidates.
Arch Phys Med Rehabil, 76, 547-551.
Bortz WM (1982) Disuse and aging. ]AMA, 248,
Bosco C, Tarrka I, Komi PV (1982) Effect of elastic
energy and myoelectrical potentiation of triceps
surae during stretch-shortening cycle exercises.
Int] Sports Med, 3,137-140.
Bottini G, Sterzi R, Vallar G (1992) Directional
hypokinesia in spatial hemineglect: a case
study. J Neurol Neurosurg Psychiatry, 55,
Bouisset S, Duchenne J-L (1994) Is body balance
more perturbed by respiration in seating
than in standing posture? NeuroReport, 5,
Bourbonnais D, Vanden Noven S (1989) Weakness
in patients with hemiparesis. Am] Occup Ther,
Bradford J, McFadyen BJ, Winter DA (1988) An
integrated biomechanical analysis of normal stair
ascent and descent.] Biomech, 21, 733-744.
Branch LG, Meyers AR (1987) Assessing physical
function in the elderly. Clin Geriatr Med, 3,
Brauer SG, Burns YR, Galley P (2000) A
prospective study of laboratory and clinical
measures of postural stability to predict
community-dwelling fallers.] Gerontol, 55A,
Brinkman JR, Hoskins TA (1979) Physical
conditioning and altered self-concept in
rehabilitated hemiplegic patients. Phys Ther, 59,
Brodal A (1973) Self-observation and
neuroanatomical considerations after a stroke.
Brain, 96,675-694.
Broeks JG, Lankhorst GJ, Rumping K et al. (1999)
The long-term outcome of arm function after
stroke: results of a follow-up study. Disabil
Rehabil, 21, 357-364.
Brown DA, Kautz SA (1998) Increased workload
enhances force output during pedalling exercise
in persons with poststroke hemiplegia. Stroke,
29, 598-606.
Brown P (1994) Pathophysiology of spasticity.
J Neurol Neurosurg Psychiatry, 57, 773-777.
Brunnstrom S (1970) Movement Therapy in
Hemiplegia: A Neurophysiological Approach.
Harper and Row, New York.
Brunt D, Liu S, Trimble Met al. (1999) Principles
underlying the organization of movement
initiation from quiet stance. Gait Posture, 10,
Buchner DM, de Lateur BJ (1991) The importance
of skeletal muscle strength to physical function
in older adults. Ann Behav Med, 13,91-98.
Buchner DM, Larson EB, Wagner EH et al. (1996)
Evidence for a non-linear relationship between
leg strength and gait speed. Age Ageing, 25,
Burdett RG, Habasevich R, Pisciotta Jet al. (1985)
Biomechanical comparison of rising from two
types of chairs. Phys Ther, 65,1177-1183.
Burdett RG, Borello-France D, Blatchly C et al.
(1988) Gait comparison of subjects with
hemiplegia walking unbraced, with ankle-foot
orthosis, and with Air-Stirrup Brace. Phys Ther,
Burke D (1988) Spasticity as an adaptation to
pyramidal tract injury. Adv Neurol, 47,
Burke D, Gandevia SC (1988) Interfering cutaneous
stimulation and the muscle afferent contribution
to cortical potentials. Electroencephalogr Clin
Neurophysiol, 70, 118-125.
Burke D, Andrews C, Ashby P (1971) Autogenic
effect of static muscle stretch in spastic man.
Arch Neurol, 25, 367-372.
Burke D, Hagbarth KE, Lofstedt L (1978) Muscle
spindle activity in man during shortening and
lengthening contractions. ] Physiol, 277,
Butefisch C, Hummelsheim H, Mauritz K-H (1995)
Repetitive training of isolated movements
improves the outcome of motor rehabilitation of
the centrally paretic hand.
] Neurol Sci, 130, 59-68.
Cahill BM, Carr JH, Adams R (1999) Inter
segmental co-ordination in sit-to-stand: an age
cross-sectional study. Physiother Res lnt, 4,
Cailliet R (1991) Shoulder Pain. FA Davis,
Calvanio R, Levine D, Petrone P (1993) Elements
of cognitive rehabilitation after right hemisphere
stroke. Behav Neurol, 11,25-57.
Campbell AJ, Robertson MC, Gardner MM et al.
(1997) Randomized controlled trial of a general
practice programme of home based exercise to
prevent falls in elderly women. BM], 315,
Canning CG, Ada L, O'Dwyer N (1999) Slowness
to develop force contributes to weakness after
stroke. Arch Phys Med Rehabil, 80, 66-70.
Canning CG, Ada L, O'Dwyer NJ (2000)
Abnormal muscle activation characteristics
associated with loss of dexterity after stroke.
] Neurol Sci, 176,45-56.
Carey JR, Burghardt TP (1993) Movement
dysfunction following central nervous system
lesions: a problem of neurologic or muscular
impairment? Phys Ther, 73, 538-547.
Carey JR, Allison JD, Mundale MO (1983)
Electromyographic study of muscular overflow
during precision handgrip. Phys Ther, 63,
Carey LM (1995) Somatosensory loss after stroke.
Crit Rev Phys Med Rehabil, 7, 51-91.
Carey LM, Matyas TA, Oke LE et al. (1993)
Sensory loss in stroke patients: effective training
of tactile and proprioceptive discrimination.
Arch Phys Med Rehabil, 74, 602-611.
Carlsoo S, Dahllof A, Holm J (1974) Kinetic
analysis of the gait in patients with hemiparesis
and in patients with intermittent claudication.
ScandJ Rehabil Med, 6, 166-179.
Carolan B, Cafarelli E (1992) Adaptations in
coactivation after isometric resistance training.
] Appl Physiol, 73, 911-917.
Carr JH, Gentile AM (1994) The effect of arm
movement on the biomechanics of standing up.
Human Mov Sci, 13, 175-193.
Carr JH, Shepherd RB (1987) A Motor Relearning
Programme for Stroke, 2nd edn, Butterworth
Heinemann, Oxford.
Carr JH, Shepherd RB (1998) Neurological
Rehabilitation Optimizing Motor Performance.
Butterworth-Heinemann, Oxford.
Carr JH, Shepherd RB (2000) A motor learning
model for rehabilitation. In Movement Science
Foundations for Physical Therapy in
Rehabilitation (eds JH Carr, RB Shepherd),
PPO-ED Austin, TX, pp 33-110.
Carr JH, Shepherd RB, Nordholm L et al. (1985)
Investigation of a new motor assessment scale
for stroke patients. Phys Ther, 65, 175-180.
Carr JH, OW JEG, Shepherd RB (2002) Some
biomechanical characteristics of standing up at
three different speeds: implications for functional
training. Physiother Theory Prac, (in press).
Castiello U, Bennett KMB, Paulignan Y (1992)
Does the type of prehension influence the
kinematics of reaching? Behav Brain Res, 50,

Castiello U, Bennett KMB, Stelmach GE (1993)
The bilateral reach to grasp movement. Behav
Brain Res, 56, 43-57.
Cauraugh J, Light K, Kim S et al. (2000) Chronic
motor dysfunction after stroke: recovering wrist
and finger extension by electromyography
triggered neuromuscular stimulation. Stroke, 31,
Chaco J, Wolf E (1971) Subluxation of the
glenohumeral joint in hemiplegia. Am] Phys
Med, 50,139-143.
Chae J, Yu D (1999) Neuromuscular stimulation

for motor relearning in hemiplegia. Crit Rev

Phys Rehabil Med, 11,279-297.
Chae J, Yu D (2000) A critical review of
neuromuscular electrical stimulation for
treatment of motor dysfunction in hemiplegia.
Asst Technol, 12,33-49.
Chae J, Bethoux F, Bohinc T et al. (1998)
Neuromuscular stimulation for upper extremity
motor and functional recovery in acute
hemiplegia. Stroke, 29,975-979.
Chantrain A, Baribeault A, Uebelhart D et al. (1999)
Shoulder pain and dysfunction in hemiplegia:
effects of functional electrical stimulation. Arch
Phys Med Rehabil, 80, 328-331.
Chapman CE, Wiesendanger M (1982) The
physiological and anatomical basis of spasticity:
a review. Physiother Can, 34, 125-136.
Chari V, Kirby RL (1986) Lower-limb influence on
sitting balance while reaching forward. Arch
Phys Med Rehabil, 67, 730-733.
Chen C, Chen H, Wong M et al. (2001) Temporal
stride and force analysis of cane-assisted gait in
people with hemiplegic gait. Arch Phys Med
Rehabil, 82,43-48.
Chen H-C, Schultz AB, Ashton-Miller JA et al.
(1996) Stepping over obstacles: dividing
attention impairs performance of old more than
young adults. ] Gerontol: Med Sci, 51A,
Cifu DX, Stewart DG (1996) A comprehensive
annotated reference guide to outcome after
stroke. Crit Rev Phys Rehab Med, 8, 39-86.
Cirstea MC, Levin MF (2000) Compensatory
strategies for reaching in stroke. Brain, 123,
Cohen LG, Roth BJ, Wasserman EM et al. (1991)
Magnetic stimulation of the human cerebral
cortex, an indicator of reorganization in motor
pathways in certain pathological conditions.
] Clin Neurophysiol, 8, 56-65.
Colebatch JG, Gandevia SC (1989) The
distribution of muscular weakness in upper
motor neuron lesions affecting the arm. Brain,
112, 749-763.
Colebatch JG, Rothwell JC, Day BL et al. (1990)
Cortical outflow to proximal arm muscles in
man. Brain, 113, 1843-1856.
Colle FM (1995) The measurement of standing
balance after stroke. Physiother Theory Pract,
Collin C, Wade D (1990) Assessing motor
impairment after stroke: a pilot reliability study.
] Neurol Neurosurg Psychiatry, 53, 576-579.
Commissaris DACM, Toussaint HM, Hirschfeld H
(2001) Anticipatory postural adjustments
in a bimanual, whole-body lifting task seem not
only aimed at minimising anterior-posterior
centre of mass displacements. Gait Posture,
Coote S, Stokes EK (2001) Physiotherapy for upper
extremity dysfunction following stroke. Phys
Ther Rev, 6, 63-69.
Cordo PJ, Nashner LM (1982) Properties of
postural adjustments associated with rapid arm
movements. ] Neurophysiol, 47, 287-302.
Coslett HB, Bowers D, Fitzpatrick E et al. (1990)
Directional hypokinesia and hemispatial
inattention in neglect. Brain, 113,475-486.
Cramer SC, Bastings EP (2000) Mapping clinically
relevant plasticity after stroke.
Neuropharmacology, 39, 842-851.
Cramer SC, Nelles G, Benson RR et al. (1997)
A functional MRI study of subjects recovered
from hemiparetic stroke. Stroke, 28, 2518-2527.
Crenna P, Frigo C, Massion Jet al. (1987) Forward
and backward axial synergies in man. Exp
Brain Res, 65, 538-548.
Critchley M (1953) The Parietal Lobes. Hafner,
New York.
Crompton S, Khemlani M, Batty Jet al. (2001)
Practical issues in retraining walking in severely
disabled patients using treadmill and harness
support systems. AustJ Physiother, 47,
Crosbie J (1993) Kinematics of walking frame
ambulation. Clin Biomech, 8, 31-36.
Crosbie J, Ko V (2000) Changes in the temporal
and distance parameters of gait evoked by
negotiation of curbs. AustJ Physiother, 46,
Crosbie J, Shepherd RB, Squire TJ (1995) Postural
and voluntary movement during reaching in
sitting: the role of the lower limbs.
] Hum Mov Stud, 28,103-126.
Crosbie J, Vachalathiti R, Smith R (1997a) Patterns
of spinal motion during walking. Gait Posture,
Crosbie J, Vachalathiti R, Smith R (1997b) Age,
gender and speed effects on spinal motion
during walking. Gait Posture, 5, 13-20.
Cruz-Martinez A (1984) Electrophysiological study
in hemiparetic subjects: electromyography,
motor conduction and response to repetitive
nerve stimulation. Electroencephalogr Clin
Neurophysiol, 23,139-148.
Culham EG, Noce RR, Bagg SD (1995) Shoulder
complex position and glenohumeral subluxation
in hemiplegia. Arch Phys Med Rehabil, 76,
Dannenbaum RM, Dykes RW (1988) Sensory loss
in the hand after sensory stroke: therapeutic
rationale. Arch Phys Med Rehabil, 69,
Danneskiold-Samsoe B, Kofod V, Munter Jet al.
(1984) Muscle strength and functional capacity
in 78-81-year-old men and women. EurJ Appl
Physiol, 52, 310-314.
Darien-Smith I, Galea MP, Darien-Smith C (1996)
Manual dexterity: how does the cerebral cortex
contribute? Clin Exp Pharmacol Physiol, 23,
Davies JM, Mayston MJ, Newham DJ (1996)
Electrical and mechanical output of the knee
muscles during isometric and isokinetic activity
in stroke and healthy subjects. Disabil Rehabil,
Davies PM (1985) Steps to Follow, A Guide to the
Treatment of Adult Hemiplegia. Springer, New
Davies PM (1990) Right in the Middle - Selective
Trunk Activity in the Treatment of Adult
Hemiplegia. Springer, Berlin.
Day BL, Steiger MJ, Thompson PD et al. (1998)
Effect of vision and stance width on human
body motion when standing: implications for
afferent control of lateral sway. ] Physiol, 469,
Dean C, Mackey F (1992) Motor assessment
scale scores as a measure of rehabilitation
outcome following stroke. AustJ Physiother,
Dean CM, Shepherd RB (1997) Task-related
training improves performance of
seated reaching tasks after stroke: a
randomized controlled trial. Stroke, 28,
Dean C, Shepherd R, Adams R (1999a) Sitting
balance I: trunk-arm and the contribution of
the lower limbs during self-paced reaching in
sitting. Gait Posture, 10, 135-144.
Dean C, Shepherd RB, Adams RD (1999b) Sitting
balance II: reach direction and thigh support
affect the contribution of the lower limbs when
reaching beyond arm's length in sitting. Gait
Posture, 10, 147-153.
Dean CM, Richards CL, Malouin F (2000)
Task-related training improves performance
of locomotor tasks in chronic stroke. A
randomized controlled pilot trial. Arch Phys
Med Rehabil, 81,409-417.
Dean CM, Richards CL, Malouin F (2001)
Walking speed over 10 metres overestimates
locomotor capacity after stroke. Clin Rehabil,
De Deyne PG (2001) Application of passive stretch
and its implications for muscle fibres. Phys
Ther, 81, 819-827.
Desrosiers J, Bourbonnais 0, Bravo G et al. (1996)
Performance of 'unaffected' upper extremity of
elderly stroke patients. Stroke, 27, 1564-1570.
Dettman MA, Linder MT, Sepic SB (1987)
Relationships among walking performance,
postural stability, and functional assessments of
the hemiplegic patient. Am I Phys Med, 66,
De Weerdt W, Spaepen A (1999) Balance. In
Functional Human Movement: Measurement
and Analysis (eds BR Durward, GO Baer,
PJ Rowe), Butterworth-Heinemann, Oxford,
pp 203-218.
De Weerdt W, Nuyens G, Feys H et al. (2001)
Group physiotherapy improves time use by
patients with stroke in rehabilitation. Aust I
Physiother, 47,53-61.
Dickstein R, Nissan M, Pillar T et al. (1984)
Foot-ground pressure pattern of standing
hemiplegic patients. Phys Ther, 64, 19-23.
Dickstein R, Heffes Y, Abulaffio N (1996)
Electromyographic and positional changes in the
elbows of spastic hemiparetic patients during
walking. Electroencephalogr Clin Neurophysiol,
Dickstein R, Heffes Y, Laufer Y et al. (1999)
Activation of selected trunk muscles during
symmetric functional activities in poststroke
hemiparetic and hemiplegic patients. I Neurol
Neurosurg Psychiatry, 66, 218-221.
Diener HC, Bacher M, Guschlbauer B et al. (1993)
The coordination of posture and voluntary
movement in patients with hemiparesis. I
Neurol, 240,161-167.
Dieterich M, Brandt T (1992) Wallenberg's
syndrome: lateropulsion, cyclorotation and
subjective visual vertical in thirty-six patients.
Ann Neurol, 31, 399-408.
Dietz V, Berger W (1983) Normal and impaired
regulation of muscle stiffness in gait: a new
hypothesis about muscle hypertonia. Exp
Neural, 79, 680-687.
Dietz V, Quintern J, Berger W (1981)
Electrophysiological studies of gait in spasticity
and rigidity. Evidence that altered mechanical
properties of muscle contribute to hypertonia.
Brain, 104,431-449.
Dietz V, Quintern J, Berger W (1984) Corrective
reactions to stumbling in man: functional
significance of spinal and transcortical reflexes.
Neurosci Lett, 44,131-135.
Dietz V, Ketelsen V-P, Berger Wet al. (1986)
Motor unit involvement in spastic paresis.
Relationship between leg muscle activation and
histochemistry. ] Neurol Sci, 75, 89-103.
di Fabio RP (1997) Adaptation of postural stability
following stroke. Topics Stroke Rehabil, 3,
di Fabio RP, Badke MB (1990) Extraneous
movement associated with hemiplegic postural
sway during dynamic goal-directed weight
redistribution. Arch Phys Med Rehabil, 71,
Dimitrijevic M, Stokic S, Wawro A et al. (1996)
Modification of motor control of wrist
extension by mesh-glove electrical afferent
stimulation in stroke patients. Arch Phys Med
Rehabil, 77,252-258.
Do MC, Bussel B, Breniere Y (1990) Influence
of plantar cutaneous afferents on early
compensatory reactions to forward fall. Exp
Brain Res, 79, 319-324.
Dobkin BH (1999) An overview of treadmill
locomotor training with partial body weight
support: a neurophysiologically sound approach
whose time has come for randomized clinical
trials. Neurorehabil Neural Repair, 13,
Dromerick AW, Edwards OF, Hahn M (2000)
Does the application of constraint-induced
movement therapy during acute rehabilitation
reduce arm impairment after ischaemic stroke?
Stroke, 31, 2984-2988.
Duncan PW (1997) Synthesis of intervention trials
to improve motor recovery following stroke.
Topics Stroke Rehabil, 3, 1-20.
Duncan PW, Weiner OK, Chandler Jet al. (1990)
Functional reach: a new clinical measure of
balance. ] Gerontol, 45, Ml92-M197.
Duncan PW, Goldstein LB, Horner RD et al.
(1994) Similar motor recovery of upper and
lower extremities after stroke. Stroke, 25,
Duncan P, Richards L, Wallace 0 et al. (1998)
A randomized controlled pilot study of a
home-based exercise program for individuals
with mild and moderate stroke. Stroke, 29,
Duong B, Low M, Moseley AM et al. (2001) Time
course of stress relaxation and recovery in
human ankles. Clin Biomech, 16, 601-607.
Dvir Z, Panturin E, Prop I (1996) The effect of
graded effort on the severity of associated
reactions in hemiplegic patients. Clin Rehabil,
Eastlake ME, Arvidson J, Snyder-Mackler L et aJ.
(1991) Interrater reliability of videotaped
observational gait-analysis assessments. Phys
Ther, 71,465-472.
Ebrahim S, Barer D, Nouri F (1986) Use of
Nottingham health profile with patients after
stroke. J Epidemiol Community Health, 40,
Edstrom L, Grimby L (1986) Effect of exercise on
the motor unit. Muscle Nerve, 9, 104-126.
Elbert T, Pantev C, Wienbruch C et al. (1995)
Increased cortical representation of the fingers
of the left hand in string players. Science, 270,
Eng JJ, Winter CD, Patla AE et al. (1992) Role of
the torque stabilizer in postural control during
rapid voluntary arm movements. In Posture and
Gait: Control Mechanisms (eds M Woollacott,
F Horak), XIth Int Symposium of Society for
Postural and Gait Research. University of
Oregon Books, Portland, OR.
Engardt M (1994) Long term effects of auditory
feedback training on relearned symmetrical
body weight distribution in stroke. A follow-up
study. Scand J Rehabil Med, 26,65-69.
Engardt M, Olsson E (1992) Body weight-bearing
while rising and sitting down in patients with
stroke. ScandJ Rehabil Med, 24, 67-74.
Engardt M, Ribbe T, Olsson E (1993) Vertical
ground reaction force feedback to enhance
stroke patients' symmetrical body-weight
distribution while rising/sitting down. Scand
J Rehabil Med, 25, 41-48.
Engardt M, Knutsson E, Jonsson M et al. (1995)
Dynamic muscle strength training in stroke
patients: effects on knee extension torque,
electromyographic activity, and motor function.
Arch Phys Med Rehabil, 76, 419-425.
Enright PL, Sherrill DL (1998) Reference equations
for the six-minute walk in healthy adults. Am J
Respir Crit Care Med, 158, 1384-1387.
Esmonde T, McGinley J, Goldie P et al. (1997)
Stroke rehabilitation: patient activity during
non-therapy time. Aust 1Physiother, 43,
Faghri PD, Rodgers MM, Glaser RM et al. (1994)
The effects of functional stimulation on
shoulder pain in hemiplegic stroke patients.
Arch Phys Med Rehabil, 75, 73-79.
Farmer SF, Swash M, Ingram DA et al. (1993)
Changes in motor unit synchronization
following central nervous lesions in man.
1Physiol, 463, 83-105.
Fellows SJ, Kaus C, Thilmann AF (1994)
Voluntary movement at the elbow in spastic
hemiparesis. Ann Neurol, 36, 397-407.
Feltz DL, Landers DM, Becker BJ (1988) A revised
meta-analysis of the mental practice literature
on motor skill learning. In Enhancing Human
Performance: Issue, Theories and Techniques
(eds D Druckman, J Swets) National Academy
Press, Washington, pp 61-101.
Fiatarone MA, Marks EC, Ryan ND et al. (1990)
High-intensity strength training in
nonagenarians. JAMA, 263, 3029-3034.
Fiatarone MA, O'Neill EF, Doyle N et al. (1993)
The Boston FICSIT study: the effects of
resistance training and nutritional
supplementation on physical frailty in the oldest
old. 1Am Ceriatr Soc, 41, 333-337.
Fiatarone MR, O'Neill EF, Ryan ND et al. (1994)
Exercise training and nutritional
supplementation for physical frailty in very
elderly people. N Engl 1Med, 330,1769-1775.
Fielding RA, Le Brasseur NK, Cuoco A et al.
(2002) High velocity power training increases
skeletal muscle peak power in older women.
lAm Geriatr Soc, 50, 655-662.
Finch L, Barbeau H, Arsenault B (1991) Influence
of body weight support on normal human gait:
development of a gait retraining strategy. Phys
Ther, 71, 842-856.
Finger S (1978) Environmental attenuation of brain
lesion symptoms. In Recovery from Brain
Damage (ed S Finger), Plenum Press, New York.
Finley FR, Cody KA (1970) locomotive
characteristics of urban pedestrians. Arch Phys
Med Rehabil, 51, 423-426.
Fisher SV, Gullickson G Jr (1978) Energy cost
of ambulation in health and disability: a
literature review. Arch Phys Med Rehabil,
59, 124-133.
Fishman MN, Colby lA, Sachs LA et al. (1997)
Comparison of upper-extremity balance tasks
and force platform testing in persons with
hemiparesis. Phys Ther, 77, 1052-1062.
Fisk JD, Goodale MA (1988) The effects of
unilateral brain damage on visually guided
reaching: hemispheric differences in the nature
of the deficit. Exp Brain Res, 72,425-435.
Fitts PM, Posner MI (1967) Human Performance.
Brooks/Cole, Belmont, CA.
Fix A, Daughton D (1988) Human Activity Profile
Professional Manual. Psychological Assessment
Resources Inc, Odessa, FL.
Fleckenstein SJ, Kirby RL, Macleod DA (1988)
Effect of limited knee-flexion range on peak hip
moments of force while transferring from sitting
to standing. J Biomech, 21, 915-918.
Fletcher GF, Schlant RC (1994) The exercise test.
In The Heart: Arteries and Veins (eds RC
Schlant, RW Alexander), McGraw-Hill,
New York, 8th edn, pp 423-440.
Foldvari M, Clark M, Laviolette lC et al. (2000)
Association of muscle power with functional
status in community-dwelling elderly women.
J Gerontol, 55A, MI92-M199.
Forssberg H (1982) Spinal locomotion functions
and descending control. In Brain Stem
Control of Spinal Mechanisms (eds B Sjolund,
A Bjorklund), Elsevier Biomedical Press,
New York.
Forster A, Young J (1995) Incidence and
consequences of falls due to stroke: a systematic
inquiry. BM], 311, 83-86.
Fowler V, Carr JH (1996) Auditory feedback:
effects on vertical force production during
standing up following stroke. Int] Rehabil Res,
Fowler V, Canning CG, Carr JH et al. (1997)
The effect of muscle length on the pendulum
test. Arch Phys Med Rehabil, 79, 169-171.
Francisco G, Chae J, Chawla H et al. (1998)
Electromyogram-triggered neuromuscular
stimulation for improving the arm function
of acute stroke survivors: a randomized
pilot study. Arch Phys Med Rehabil, 79,
Friel KM, Nudo RJ (1998) Restraint of the
unimpaired hand is not sufficient to retain
spared hand representation after focal cortical
injury. Soc Neurosci Abst, 24: 405.
Frontera WR, Meredith CN, O'Reilly KP et al.
(1988) Strength conditioning in older men:
skeletal muscle hypertrophy and improved
function.] Appl Physiol, 64, 1038-1044.
Fuhr P, Cohen LG, Dang Net al. (1992)
Physiological analysis of motor reorganization
following lower limb amputation.
Electroencephalogr Clin Neurophysiol,
Furlan M, Marchal G, Vialder F et al. (1996)
Spontaneous neurological recovery after stroke
and the fate of the ischemic penumbra. Ann
Neurol, 40, 216-226.
Gabell A, Nayak VSC (1984) The effect of age on
variability in gait.] Gerontol, 39, 662-666.
Gabell A, Simons MA, Mayak USL (1986) Falls in
the healthy elderly: predisposing causes.
Ergonomics, 28, 965-975.
Gallistel CR (1980) The Organization of Action:
A New Synthesis. Lawrence Erlbaum,
Hillsdale. NJ.
Gardner MM, Robertson MC, Campbell AJ
(2000) Exercise in preventing falls and fall
related injuries in older people: a review of
randomized controlled trials. Br ] Sports Med,
Gatev P, Thomas S, Kepple T et al. (1999)
Feedforward strategy of balance during quiet
stance in adults.] Physiol, 514, 915-928.
Gehlsen GA, Whaley MH (1990) Falls in the
elderly: part II, balance, strength, and flexibility.
Arch Phys Med Rehabil, 71, 739-741.
Gelber DA, Good DC, Laven LJ et al. (1993)
Causes of urinary incontinence after stroke.
Stroke, 24, 378-382.
Gemperline 11, Allen S, Walk D et al. (1995)
Characteristics of motor unit discharge in
subjects with hemiparesis. Muscle Nerve, 18,
Gentile AM (1987) Skill acquisition: action,
movement, and neuromotor processes. In
Movement Science Foundations for Physical
Therapy in Rehabilitation (eds JH Carr,
RB Shepherd), Aspen Publishers,
Rockville, MD, pp 93-154.
Gentile AM (2000) Skill acquisition: action,
movement, and neuromotor processes. In
Movement Science Foundations for Physical
Therapy in Rehabilitation (eds JH Carr,
RB Shepherd), PRO-ED Austin, TX, 2nd edn,
pp 111-187.
Ghez C (1991) Posture. In Principles of Neural
Science (eds ER Kandell, JH Schwartz, TM
Jessell), Appleton and Lange, Norwalk, CT, 3rd
edn, pp 596-608.
Gibson 11 (1977) The theory of affordances.
In Perceiving, Action and Knowing: Towards
an Ecological Psychology (eds R Shaw,
J Bransford), Erlbaum, Hillsdale, NJ, pp 67-82.
Gill-Body K, Krebs D (1994) Usefulness of
biomechanical measurements approaches in
rehabilitation. Topics Geriatr Rehabil, 10,
Gioux M, Petit J (1993) Effects of immobilizing
the cat peroneus longus muscle on the
activity of its own spindles.] Appl Physiol,
Glanz M, Klawansky S, Stason W et al. (1996)
Functional electrostimulation in poststroke
rehabilitation: a meta-analysis of the
randomized controlled trials. Arch Phys Med
Rehabil, 77, 549-553.
Goldie P, Matyas T, Kinsella G (1992) Movement
rehabilitation following stroke. Research
Report, Department of Health, Housing and
Community Services, Victoria, Australia.
Goldie PA, Matyas TA, Evans OM et al. (1996a)
Maximum voluntary weight-bearing by the
affected and unaffected legs in standing
following stroke. Clin Biomech, 11, 333-342.
Goldie PA, Evans 0, Matyas T (1996b)
Performance in the stability limits test during
rehabilitation following stroke. Gait Posture, 4,
Goodale MA, Milner AD, Jakobson LS et al.
(1990) Kinematic analysis of limb movements in
neuropsychological research: subtle deficits and
recovery of function. Can] Psychol, 44,
Gordon J (2000) Assumptions underlying
physical therapy interventions: theoretical
and historical perspectives. In Movement
Science Foundations for Physical Therapy in
Rehabilitation (eds JH Carr, RB Shepherd),
Aspen Publishers, Rockville, MD, 2nd edn,
pp 1-31.
Gottlieb GL, Corcos DM, Jaric S et al. (1988)
Practice improves even the simplest movements.
Exp Brain Res, 73, 436-440.
Gowland C (1982) Recovery of motor function
following stroke: profile and predictors.
Physiother Can, 34, 77-84.
Gowland C, deBruin H, Basmajian JV et al. (1992)
Agonist and antagonist activity during voluntary
upper-limb movement in patients with stroke.
Phys Ther, 72, 624-633.
Grabiner MD, Koh TJ, Lundin TM et al. (1993)
Kinematics of recovery of a stumble. J Gerontol,
3, M97-MI02.
Gracies J-M, Wilson L, Gandevia SC et al. (1997)
Stretched position of spastic muscles aggravates
their co-contraction in hemiplegic patients.
Ann Neurol, 42, 438-439.
Gracies J-M, Marosszeky JE, Renton R et al.
(2000) Short term effects of dynamic Iycra
splints on upper limb in hemiplegic patients.
Arch Phys Med Rehabil, 81, 1547-1555.
Granger CV, Hamilton BB, Sherwin FS (1986)
Guide for the Use of the Uniform Data Set for
Medical Rehabilitation. Project Office, Buffalo
General Hospital, New York.
Gregory JE, Morgan DL, Proske U (1988)
After-effects in the responses of cat muscle
spindles and errors in limb position sense in
man. J Neurophysiol, 59, 1220-1230.
Gresham G, Duncan P, Stason W et al. (1995)
Post-stroke rehabilitation: clinical practice
guidelines, no 16. US Department of Health and
Human Services, Public Health Service, AHCPR
95-0662, Rockville, MD.
Griffin MP, Olney SJ, McBride ID (1995) Role of
symmetry in gait performance of stroke subjects
with hemiplegia. Gait Posture, 3, 132-142.
Grillner S, Shik ML (1973) On the descending
control of the lumbosacral spinal cord from
'mesencephalic locomotor region'. Acta Physiol
Scand, 87, 320-333.
Grimby G (1983) On the energy cost of achieving
mobility. Scand J Rehabil Med Suppl, 9, 49-54.
Guiliani CA (1990) Adult hemiplegic gait. In Gait
in Rehabilitation (ed G Smidt), Churchill
Livingstone, New York, pp 253-266.
Guyatt GH, Pugsley SO, Sullivan MJ et al. (1984)
Effect of encouragement on walking test
performance. Thorax, 39, 818-822.
Guyatt GH, Sullivan MJ, Thompson pJ et al.
(1985) The 6-minute walk: a new measure of
exercise capacity in patients with chronic heart
failure. Can Med J, 132,919-923.
Hachisuka K, Umezu Y, Ogata B (1997) Disuse
muscle atrophy of lower limbs in hemiplegic
patients. Arch Phys Med Rehabil, 78,13-18.
Hagbarth K-E, Hagglund lV, Norkin M et al.
(1985) Thixotropic behaviour of human finger
flexor muscles with accompanying changes in
spindle and reflex responses to stretch. J Physiol,
368, 323-342.
Hakkinen K, Komi PV (1981) Effect of different
combined concentric and eccentric muscle work
regimens on maximal strength development.
J Hum Mov Stud, 7, 33-44.
Hakkinen K, Komi PV (1983) Electromyographic
changes during strength training and detraining.
Med Sci Sports Exercise, 15, 455-460.
Hakkinen K, Newton RU, Gordon SE et al. (1998)
Changes in muscle morphology,
electromyographic activity, and force
production characteristics during progressive
strength training in young and older men.
JGerontol BioI Sci, 53A, B415-B423.
Halbertsma JP, van Bolhuis AI, Goeken LN (1996)
Sports stretching: effect on passive muscle
stiffness of short hamstrings. Arch Phys Med
Rehabil, 77, 688-692.
Halkjaer-Kristensen J, Ingemann-Hansen T (1985)
Wasting of the human quadriceps muscle after
knee ligament injuries. 1. Anthropometric
consequences. Scand J Rehabil Med Suppl,
Hall EJ, Flament D, Fraser C et al. (1990) Non
invasive brain stimulation reveals reorganized
cerebral output in amputees. Neurosci Lett,
116, 379-386.
Halligan PW, MarshallJC (1992) Left visuo-spatial
neglect: a meaningless entity? Cortex,
Hamdorf P, Withers R, Penhall Ret al. (1992)
Physical training effects on the fitness and
habitual activity patterns of elderly women.
Arch Phys Med Rehabil, 73, 603-608.
Hamm LF, Leon AS (1994) Exercise training for
the coronary patient. In Rehabilitation of the
Coronary Patient (eds NK Wenger, HK
Hellerstein), Churchill Livingstone, New York.
Hammond MC, Kraft GH, Fitts SS (1988)
Recruitment and termination of
electromyographic activity in the hemiparetic
forearm. Arch Phys Med Rehabil, 69,106-110.
Hamrin E, Eklund G, Hillgren A-K et al. (1982)
Muscle strength and balance in post-stroke
patients. Uppsala Med Sci, 87, 11-26.
Harlaar J, Becher JG, Snijders CJ et al. (2000)
Passive stiffness characteristics of ankle plantar
flexors in hemiplegia. Clin Biomech, 15,
Harridge SDR, Kryger A, Stensgaard A (1999)
Knee extensor strength, activation, and size in
very elderly people following strength training.
Muscle Nerve, 22,831-839.
Harris ML, Polkey MI (2001) Quadriceps muscle
weakness following acute hemiplegic stroke.
Clin Rehabil, 15,274-281.
Harvey LA, Herbert RD (2002) Muscle stretching
for treatment and prevention of contracture in
~ . _ ._XiQ_---
people with spinal cord injury. Spinal Cord,
40, 1-9.
Hase K, Stein RB (1999) Turning strategies
during human walking. J Neurophysiol, 81,
Hauptmann B, Hummelsheim H (1996)
Facilitation of motor evoked potentials in hand
extensor muscles of stroke patients: correlation
to the level of voluntary contraction.
Electroencephalogr Clin Neurophysiol, 101,
Hausdorff JM, Rios DA, Edelberg HK (2001) Gait
variability and fall risk in community-living
older adults: a I-year prospective study. Arch
Phys Med Rehabil, 82, 1050-1056.
Hazlewood ME, Brown JK, Rowe PJ et al. (1994)
The use of therapeutic electrical stimulation in
the treatment of hemiplegic cerebral palsy. Dev
Med Child Neurol, 36, 661-673.
Heilman KM, van den Abell T (1980) The right
hemisphere dominance for attention: the
mechanism underlying hemispheric asymmetries
of inattention (neglect). Neurology, 30, 327-330.
Heilman KM, Bowers D, Valenstein E et al. (1987)
Hemispace and hemispatial neglect. In
Neurophysiological and Neuropsychological
Aspects of Spatial Neglect (ed M Jeannerod),
Elsevier, Amsterdam, pp 115-150.
Held JM, Gordon J, Gentile AM (1985)
Environmental influences on locomotor recovery
following cortical lesions in rats. Behav
Neurosci, 99, 678-690.
Herbert RD, Balnave RJ (1993) The effect of
position of immobilization on resting length,
resting stiffness, and weight of the soleus muscle
of the rabbit. J Anat, 116,45-55.
Herbert EP, Landin D (1994) Effects of a
learning model and augmented feedback on
tennis skill acquisition. Res Q Exercise Sport,
Herbert R (1988) The passive mechanical
properties of muscle and their adaptations to
altered patterns of use. Aust J Physiother, 34,
Herbert RD, Sherrington C, Maher C et al. (2001)
Evidence-based practice - imperfect but
necessary. Physiother Theory Prac, 17, 201-21l.
Herman EWM (1992) Spatial neglect: new issues
and their implications for occupational therapy
practice. Am J Occup Ther, 46, 207-216.
Hermsdorfer J, Mai N (1996) Disturbed
grip-force control following cerebral lesions.
J Hand Ther, 9, 33-40.
Herzog W, Hasler E, Abrahams SK (l991a) A
comparison of knee extensor strength curves
obtained theoretically and experimentally. Med
Sci Sport Exercise, 23, 108-114.
Herzog W, Koh T, Hasler E et al. (1991b)
Specificity and plasticity of mammalian skeleton
muscles. J Appl Biomech, 16, 98-109.
Hesse S (1999) Treadmill training with partial
body weight support in hemiparetic patients
further research needed. Neurorehabil Neural
Repair, 13, 179-18l.
Hesse S, Schauer M, Malezic M et al. (1994)
Quantitative analysis of rising from a chair in
healthy and hemiparetic subjects. Scand
J Rehabil Med, 26,161-166.
Hesse S, Bertelt C, Jahnke MT et al. (1995)
Treadmill training with partial body weight
support compared with physiotherapy in
nonambulatory hemiparetic patients. Stroke,
Hesse S, Helm B, Krajnik Jet al. (1997) Treadmill
training with partial body weight support:
influence of body weight release on the gait of
hemiparetic patients. J Neurol Rehabil, 11,
Hesse S, Schauer M, Petersen M et al. (1998)
Sit-to-stand manoeuvre in hemiparetic patients
before and after a 4-week rehabilitation
programme. Scand J Rehabil Med, 30, 81-86.
Hewson L (1996) The Stroke Jigsaw: Lorna's
Story. An Integrated Resource Package.
University of Newcastle, Newcastle,
Hill KM, Harburn KL, Kramer JF et al. (1994)
Comparison of balance responses to an
external perturbation test, with and without an
overhead harness safety system. Gait Posture,
Hill K, Bernhardt J, McGann A et al. (1996)
A new test of dynamic standing balance for
stroke patients: reliability and comparison
with healthy elderly. Physiother Can,
48, 257-262.
Hill K, Ellis P, Bernhardt Jet al. (1997) Balance
and mobility outcomes for stroke patients: a
comprehensive audit. Aust J Physiother, 43,
Hintermeister RA, Bey MJ, Lange GW et al. (1998)
Quantification of elastic resistance knee
rehabilitation exercises. J Orthop Sports Phys
Ther, 28, 40-50.
Hoff B, Arbib MA (1993) Models of trajectory
formation and temporal interaction of reach and
grasp. J Motor Behav, 25, 175-192.
Hogan N, Winters JM (1990) Principles
underlying movement organization: upper limb.
In Multiple Muscle Systems: Biomechanics
and Movement Organization (eds JM Winters,
SL-Y Woo), Springer, New York, pp 182-194.
Holden MK, Gill KM, Magliozzi MR (1986)
Gait assessment for neurologically impaired
patients: standards for outcome assessment.
Phys Ther, 66,1530-1539.
Holt RR, Simpson D, Jenner JR et al. (2000)
Ground reaction force after a sideways push as
a measure of balance in recovery from stroke.
Clin Rehabil, 14, 88-95.
Horak F (1987) Clinical measurement of postural
control in adults. Phys Ther, 67,1881-1885.
Horak F (1997) Invited commentary. Phys Ther,
77, 382-383.
Horak FB, Nashner LM (1986) Central
programming of postural movements:
adaptation to altered support-surface
configuration. J Neurophysiol, 55, 1369-1381.
Horak FB, Esselman ME, Anderson ME et al.
(1984) The effects of movement velocity, mass
displaced, and task certainty on associated
postural adjustments made by normal and
hemiplegic individuals. J Neurol Neurosurg
Psychiatry, 48,1020-1028.
Horak FB, Shupert CL, Mirka A (1989)
Components of postural dyscontrol in the
elderly: a review. Neurobiol Ageing, 10,
Horak FB, Henry SM, Shumway-Cook A (1997)
Postural perturbations: new insights for
treatment of balance disorders. Phys Ther, 77,
Hortobagyi T, Tunnel D, Moody Jet al. (2001)
Low- or high-intensity strength training partially
restores impaired quadriceps force accuracy and
steadiness in aged adults. J Gerontol: BioI Sci,
56A, B38-B47.
Hu M-H, Woollacott MH (1994) Multisensory
training of standing balance in older adults:
1. Postural stability and one-leg stance balance.
J Gerontol: Med Sci, 49, M52-M61.
Hufschmidt A, Mauritz K-H (1985) Chronic
transformation of muscle in spasticity: a
peripheral contribution to increased tone.
J Neurol Neurosurg Psychiatry, 48, 676-685.
Hummelsheim H, Mauritz K-H (1993)
Neurophysiological mechanisms of spasticity
modification by physiotherapy. In
Spasticity: Mechanisms and Management (eds
AF Thilmann, DJ Burke, WZ Rymer), Springer,
New York.
Hummelsheim H, Munch B, Butefisch C et al.
(1994) Influence of sustained stretch on late
muscular response to magnetic brain stimulation
in patients with upper motor neuron lesions.
Scand J Rehabil Med, 26, 3-9.
Hummelsheim H, Maier-Loth ML, Eickhof C
(1997) The functional value of electrical
muscle stimulation for the rehabilitation of
the hand in stroke patients. Scand J Rehabil
Med, 29, 3-10.
Humphreys GW, Riddoch MJ (1992) Interactions
between object- and space-vision revealed
through neuropsychology. In Attention and
Performance IX (eds DE Meyer, S Kornblum),
Lawrence Erlbaum, Hove.
Hunt SM, McEwen J, McKenna SP (1985)
Measuring health status: a new tool for
clinicians and epidemiologists. J Royal Coll Gen
Pract, 35, 185-188.
Huxham FE, Goldie PA, Patla AE (2001)
Theoretical considerations in balance
assessment. Aust J Physiother, 47,
Hyvarinen J, Poranen A, Jokinen Y (1980)
Influence of attentive behavior on neuronal
responses to vibration in primary somatosensory
cortex of the monkey. J Neurophysiol, 43,
Iberall T, Bingham G, Arbib MA (1986)
Opposition space as a structuring concept for
the analysis of skilled hand movements. Brain
Res, 15, 158-173.
Ibrahim IK, Berger W, Trippel M et al. (1993)
Stretch-induced electromyographic activity and
torque in spastic elbow muscles. Brain, 116,
Ikai T, Tei K, Yoshida K et al. (1998) Evaluation
and treatment of shoulder subluxation in
hemiplegia. Am J Phys Med Rehabil, 77,
Ikeda ER, Schenkman ML, O'Riley P et al. (1991)
Influence of age on dynamics of rising from a
chair. Phys Ther, 71, 473-481.
Indredavik B, Slordahl SA, Bakke F et al. (1997)
Stroke unit treatment long-term effects. Stroke,
Ingen Schenau GJ, Bobbert MF, Rozendal RH
(1987) The unique action of bi-articular muscles
in complex movements. J Anat, 155, 1-5.
JacksonJH (1958) Selected writings. In John
Hughlings Jackson (ed J Taylor), Basic Books,
New York.
Janelle CM, Barba DA, Frehlich SG et al. (1997)
Maximizing performance effectiveness through
videotape replay and a self-controlled learning
environment. Res Q Exercise Sport, 68,
Jankowski LW, Sullivan 5J (1990) Aerobic and
neuromuscular training: effect on the capacity,
efficiency and fatigability of patients with
traumatic brain injuries. Arch Phys Med
Rehabil, 71, 500-504.
Jeannerod M (1981) Intersegmental coordination
during reaching at natural visual objects.
In Attention and Performance (eds J Long,
A Baddeley), vol 9, Erlbaum, Hillsdale, NJ,
pp 153-168.
Jeannerod M (1988) Neural and Behavioural
Organization of Goal-directed Movements.
Oxford University Press, Oxford.
Jeannerod M, Michel F, Prablanc C (1984) The
control of hand movements in a case of
hemianaesthesia following a parietal lesion.
Brain, 107, 899-920.
Jenkins WM, Merzenich MM, Ochs MT et al.
(1990) Functional reorganization of primary
somatosensory cortex in adult owl monkeys
after behaviorally controlled tactile stimulation.
J Neurophysiol, 68,82-104.
Jenner JR, Kirker S, Simpson D et al. (1997)
Sideways balance after stroke - a serial study
of hip abductors and adductors. Eur J Neurol,
+Johansson BB (2000) Brain plasticity and stroke
rehabilitation: the Willis Lecture. Stroke, 31,
Johansson G, Jarnlo G-B (1991) Balance training in
70-year-old women. Physiother Theory Pract, 7,
Johansson RS, Westling G (1984) Roles of
glabrous skin receptors and sensorimotor
memory in automatic control of precision grip
when lifting rougher or more slippery objects.
Exp Brain Res, 56, 550-564.
Johansson RS, Westling G (1988) Programmed and
triggered actions to rapid load changes during
precision grip. Exp Brain Res, 71, 72-86.
Johansson RS, Westling G (1990) Tactile afferent
signals in the control of precision grip. In
Attention and Performance (ed M Jeannerod),
Erlbaum, Hillsdale, NJ, pp 677-713.
Johansson RS, Riso R, Hager C et al. (1992a)
Somatosensory control of precision grip
during unpredictable pulling loads. I. Changes
in load force amplitude. Exp Brain Res, 89,
Johansson RS, Hager C, Riso R et al. (1992b)
Somatosensory control of precision grip during
unpredictable pulling loads. II. Changes in load
force rate. Exp Brain Res, 89, 192-203.
Johansson RS, Hager C, Backstrom L (1992c)
Somatosensory control of precision grip during
unpredictable pulling loads. III. Impairments
during digital anesthesia. Exp Brain Res, 89,
Joyce BM, Kirby RL (1991) Canes, crutches and
walkers. Am Fam Physician, 43, 535-542.
Judge JO, Underwood M, Gennosa T (1993)
Exercise to improve gait velocity in older
persons. Arch Phys Med Rehabil, 74,400-406.
Kaminski TR, Bock C, Gentile AM (1995) The
coordination between trunk and arm motion
during pointing movements. Exp Brain Res,
106, 457-466.
Kapanji AI (1992) Clinical evaluation of the
thumb's opposition. J Hand Ther, 2, 102-106.
Karnath HO, Schenkel P, Fischer B (1991) Trunk
orientation as the determining factor of the
'contralateral' deficit in the neglect syndrome
and as the physical anchor of the internal
representation of body orientation in space.
Brain, 114, 1997-2014.
Kautz SA, Brown DA (1998) Relationships
between timing of muscle excitation and
impaired motor performance during cyclical
lower extremity movement in post-stroke
hemiplegia. Brain, 121,515-526.
Kay TM, Myers AM, Huijbregts MPJ (2001) How
far have we come since 1992? A comparative
survey of physiotherapists' use of outcome
measures. Physiother Can, Fall, 268-275.
Keith RA (1980) Activity patterns in a stroke
rehabilitation unit. Soc Sci & Med, 14A,
Keith RA (1995) Conceptual basis of outcome
measures. Am J Phys Med & Rehabil, 74,
Keith RA, Cowell KS (1987) Time use of stroke
patients in three rehabilitation hospitals.
Soc Sci & Med, 24, 529-533.
Kelley DL, Dainis A, Wood GK (1976) Mechanics
and muscular dynamics of rising from a seated
position. In Biomechanics V-B (ed PV Komi),
University Park Press, Baltimore, pp 127-134.
Kelly J (2002) Cardiorespiratory fitness and
walking ability in acute stroke patients.
Master thesis.
Kelso JAS, Buchanan J], Murata T (1994)
Multifunctionality and switching in the
coordination dynamics of reaching and
grasping. Hum Mov Sci, 13, 63-94.
Kenney WL, Humphrey RH, Bryant CX (eds)
(1995) ACSM's Guidelines for Exercise Testing
and Prescription. Williams & Wilkins,
Philadelphia, 5th edn.
Khemlani MM, Carr JH, Crosbie WJ (1998)
Muscle synergies and joint linkages in sit-to
stand under two initial foot positions. Clin
Biomech, 14,236-246.
Kilbreath SL, Gandevia SC (1994) Limited
independent flexion of the thumb and fingers in
human subjects. J Physiol, 479, 487-497.
King TI (1993) Hand strengthening with a
computer for purposeful activity. Am J Occup
Ther, 47,635-637.
Kinsella G, Ford B (1980) Acute recovery patterns
in stroke patients. Neuropsychological factors.
Med fAust, 2, 663-666.
Kirby RL, Price NA, MacLeod DA (1987) The
influence of foot position on standing balance.
J Biomech, 20, 423-427.
Kirker SGB, Simpson S, Jenner JR et al. (2000)
Stepping before standing: hip muscle function
in stepping and standing balance after stroke.
J Neurol Neurosurg Psychiatry, 68, 458-464.
Kirsteins AE, Dietz F, Hwang S (1991) Evaluating
the safety and potential use of a weight-bearing
exercise, Tai-Chi Chuan, for rheumatoid
arthritis patients. Am J Phys Med Rehabil, 70,
Knutsson E, Richards C (1979) Different types of
distributed motor control in gait of hemiparetic
patients. Brain, 102, 404-430.
Koceja DM, Allway D, Earles DR (1999) Age
differences in postural sway during volitional
head movement. Arch Phys Med Rehabil, 80,
Kolb B, Gibb R (1991) Environmental enrichment
and cortical injury: behavioral and anatomical
consequences of frontal cortex lesions. Cerebral
Cortex, 1,189-198.
Komi PV (1973) Relationships between muscle
tension, EMG and velocity of contraction
under concentric and eccentric work. In New
Developments in Electromyography and
Clinical Neurophysiology (ed JE Desmedt),
Karger, Basel, pp 596-606.
Komi PV (1986) The stretch-shortening cycle and
human power output. In Human Muscle Power
(eds NC Jones et al.), Human Kinetics
Publishers, Champaign, IL.
Komi PV, Buskirk ER (1972) Effects of eccentric
and concentric muscle conditioning on tension
and electrical activity of human muscle.
Ergonomics, 15,417-434.
Kopp B, Kunkel A, Flor H et al. (1997) The Arm
Motor Ability Test: reliability, validity, and
sensitivity to change of an instrument for
assessing disabilities in activities of daily living.
Arch Phys Med Rehabil, 78, 615-620.
Kopp B, Kunkel A, Muhlnickel W et al. (1999)
Plasticity in the motor system related to therapy
induced improvement of movement after stroke.
Clin Neurosci, 4, 807-810.
Kornecki S, Kebel A, Siemienski S (2001) Muscular
co-operation during joint stabilisation, as
reflected by EMG. Eur JAppl Physiol, 84,
Kotake T, Dohi N, Kajiwara T et al. (1993) An
analysis of sit-to-stand movement. Arch Phys
Med Rehabil, 74, 1095-1099.
Kralj A, Jaeger RJ, Munih M (1990) Analysis of
standing up and sitting down in humans:
definitions and normative data presentation.
JBiomech, 23, 1123-1138.
Krebs DE, Wong D, Jevsevar D (1992) Trunk
kinematics during locomotor activities. Phys
Ther, 72, 505-514.
Krebs DE, Jette AM, Assmann SF (1998) Moderate
exercise improves gait stability in disabled
elders. Arch Phys Med Rehabil, 79, 1489-1495.
Krichevets AN, Sirokina EB, Yevsevicheva IV
et al. (1995) Computer games as a means of
movement rehabilitation. Disabil Rehabil, 17,
Kuan T, Tsou J, Fang-Chin S (1999) Hemiplegic
gait of stroke patients: the effect of using a cane.
Arch Phys Med Rehabil, 80, 777-784.
Kukulka CG, Beckman SM, Holte JB et al. (1986)
Effects of intermittent tendon pressure on alpha
motor neurone excitability. Phys Ther, 66,
Kumar R, Matter EJ, Mehta AJ et al. (1990)
Shoulder pain in hemiplegia. Am J Phys Med
Rehabil, 69, 205-208.
Kunkel A, Kopp B, Muller G et al. (1999)
Constraint-induced movement therapy for
motor recovery in chronic stroke patients. Arch
Phys Med Rehabil, 80, 624-628.
Kuo AD, Zajac FE (1993) A biomechanical
analysis of muscle strength as a limiting factor
in standing posture. J Biomech, 26(suppl 1),
Kuster M, Sakurai S, Wood GA (1995) Kinematic
and kinetic comparison of downhill and level
walking. Clinical Biomech, 10, 79-84.
Kuypers HGJM (1973) The anatomical
organization of the descending pathways and
their contributions to motor control especially
in primates. In New Developments in
Electromyography and Clinical
Neurophysiology 3 (ed JE Desmedt), Karger,
Basel, pp 38-68.
Kuzoffsky A (1990) Sensory Function and
Recovery after Stroke. Karolinska Institute,
Kwakkel G, Wagenaar RC, Koelman TW et al.
(1997) Effects of intensity of rehabilitation after
stroke: a research synthesis. Stroke, 28,
Kwakkel G, Wagenaar RC, Twisk JWR et al.
(1999) Intensity of arm training after primary
middle-cerebral-artery stroke: a randomised
trial. Lancet, 354, 191-196.
Ladavas E, Del Pesce M, Provinciali L (1989)
Unilateral attention deficits and hemispheric
asymmetries in the control of visual attention.
Neuropsychologia, 27, 353-366.
Ladavas E, Petronio A, Umilta C et al. (1990)
The deployment of visual attention in the intact
field of hemineglect patients. Cortex, 26,
Ladavas E, Menghini G, Umilta C (1994) On the
rehabilitation of hemispatial neglect. In
Cognitive Neuropsychology and Cognitive
Rehabilitation (eds MJ Riddoch, GW
Humphreys), Erlbaum, London, pp 151-172.
Lance JM (1980) Symposium synopsis. In
Spasticity: Disorder of Motor Control (eds
RG Feldman, RR Young, WP Koella), Year
Book, Chicago, pp 485-494.
Landau WM (1980) Spasticity: What is it? What is
it not? In Spasticity: Disorder Motor Control
(eds RG Feldman, RR Young, WP Koella), Year
Book, Chicago, pp 17-24.
Landau WM (1988) Parables of palsy, pills, and PT
pedagogy: a spastic dialectic. Neurology, 38,
Landin S, Hagenfeldt L, Saltin B et al. (1977)
Muscle metabolism during exercise in
hemiparetic patients. Clin Sci Mol Med, 53,
Latash LP, Latash ML (1994) A new book by N.A.
Bernstein: 'On Dexterity and its Development'.
JMot Behav, 26, 56-62.
Laufer Y, Dickstein R, Resnik S et al. (2000)
Weight-bearing shifts of hemiparetic and
healthy adults upon stepping on stairs of
various heights. Clin Rehabil, 14, 125-129.
Ledin T, Kronhed AC, Moller C et al. (1991)
Effects of balance training in elderly evaluated
by clinical tests and dynamic posturography.
] Vestibular Res, 1, 129-138.
Lee ON, Aronson E (1974) Visual proprioceptive
control of standing in human infants. Percept
Psychoanalysis, 15, 529-532.
Lee KC, Soderberg GL (1981) Relationship
between perception of joint position sense and
limb synergies in patients with hemiplegia. Phys
Ther, 10, 1433-1437.
Lee RG, van Donkelaar P (1995) Mechanisms
underlying functional recovery following stroke.
Can} Neural Sci, 22, 257-263.
Lee RG, Tonolli I, Viallet F et al. (1995)
Preparatory postural adjustments in
Parkinsonian patients with postural instability.
Can} Neurol Sci, 22,126-135.
Lee WA (1980) Anticipatory control of postural
and task muscles during rapid arm flexion.
} Motor Behav, 12, 185-196.
Lee WA, Deming L, Sahgal V (1988) Quantitative
and clinical measures of static standing balance
in hemiparetic and normal subjects. Phys Ther,
Lehman ]F, Condon SM, Price R et al. (1987) Gait
abnormalities in hemiplegia: their correction by
ankle-foot orthoses. Arch Phys Med Rehabil,
68, 763-771.
Lemon RN, Bennett KM, Werner W (1991) The
cortico-motor substrate for skilled movements
of the primate hand. In Tutorials in Motor
Neuroscience (eds] Requin, GE Stelmark),
Kluwer, Dordrecht, pp 477-495.
Leont'ev AN, Zaporozhets AV (1960)
Rehabilitation of Hand Function. Pergamon
Press, London.
Levin MF (1996) Interjoint coordination during
pointing movements is disrupted in spastic
hemiparesis. Brain, 119,281-293.
Lexell J (2000) Muscle structure and function in
chronic neurological disorders: the potential of
exercise to improve activities of daily living.
Exercise Sports Sci Rev, 28, 80-84.
Li ZM, Latash ML, Newell KM et al. (1998a)
Motor redundancy during maximal voluntary
contraction in four-finger tasks. Exp Brain Res,
Li ZM, Latash ML, Zatsiorsky VM (1998b) Force
sharing among fingers as a model of the
redundancy problem. Exp Brain Res, 119,
Lieber RL (1988) Comparison between animal and
human studies of skeletal muscle adaptation to
chronic stimulation. Clin Orthop Relat Res,
233, 19-24.
Lieber RL (1992) Skeletal Muscle and Function.
Williams & Wilkins, Baltimore, MD.
Liepert], Tegenthoff M, Malin H' (1995) Changes
in cortical motor area size during
immobilization. Electroencephalogr Clin
Neurophysiol, 97, 382-386.
Liepert], Miltner WHR, Bauder H et al.
(1998) Motor cortex plasticity during
constraint-induced movement therapy in stroke
patients. Neurasci Lett, 250, 5-8.
Liepert], Bauder H, Miltner W et al. (2000)
Treatment-induced cortical reorganization after
stroke in humans. Stroke, 31, 1210-1216.
Lieperr], Uhde I, Graf Set al. (2001) Motor cortex
plasticity during forced-use therapy in stroke
patients: a preliminary study. ] Neural, 248,
Lincoln NB, Gamlen R, Thomason H (1989)
Behavioural mapping of patients in a stroke
unit. lnt Dis Stud, 11, 149-154.
Lincoln NB, Jackson ]M, Adams SA (1998)
Reliability and revision of the Nottingham
Sensory Assessment for stroke patients.
Physiotherapy, 84, 358-365.
Lincoln NB, Parry RH, Vass CD (1999)
Randomized, controlled trial to evaluate
increased intensity of physiotherapy treatment of
arm function after stroke. Stroke, 30, 573-579.
Lindboe CF, Platou CS (1984) Effect of
immobilization of short duration on the muscle
fibre size. Clin Physiol, 4,183-188.
Lindmark B, Hamrin F (1995) Relation between
gait speed, knee muscle torque and motor scores
in post-stroke patients. Scand J Caring Sci, 9,
Lipshits MI, Mauritz K, Popov KE (1981)
Quantitative analysis of anticipatory
components of a complex voluntary movement.
Hum Physiol, 7, 165-173.
Loewen SC, Anderson BA (1990) Predictors of
stroke outcome using objective measurement
scales. Stroke, 21, 78-81.
Lord S, Castell S (1994) Physical activity program
for older persons. Effect on balance strength,
neuromuscular control and reaction time. Arch
Phys Med Rehabil, 78, 208-212.
Lord SR, Clark RD, Webster IW (1991) Postural
stability and associated physiological factors in
a population of aged persons. J Gerontal, 46,
Lundvik GA, Gard G, Salford E et al. (1999)
Interaction between patient and physiotherapist:
a qualitative study reflecting the
physiotherapist's perspective. Physiother Res
lnt, 4,89-109.
Lyngberg K, Danneskiold-Samsoe B, Halskov 0
(1988) The effect of physical training on
patients with rheumatoid arthritis: changes
in disease activity, muscle strength and
aerobic capacity. Clin Exp Rheumatol,
McAuley E, Rudolph D (1995) Physical activity,
aging, and psychological well-being. JAging
Phys Activity, 3, 67-96.
McComas AJ (1994) Human neuromuscular
adaptations that accompany changes in activity.
Med Sci Sports Exercise, 26, 1498-1509.
McComas AJ, Miller RG, Gandevia SG (1995)
Fatigue brought on malfunction of the central
and peripheral nervous systems. In Fatigue
(eds SG Gandevia et al.), Plenum Press,
New York.
McCrum R (1998) My Year Off: Rediscovering
Life after Stroke. Picador, London.
McGavin CR, Gupta SP, McHardy GJ (1976)
Twelve-minute walking test for assessing
disability in chronic bronchitis. BM], 1,
Mackenzie-Knapp M (1999) Electrical stimulation
in early stroke rehabilitation of the upper limb
within attention. Aust] Physiother, 45,
MacKinnon CD, Winter DA (1993) Control of
whole body balance in the frontal plane during
human walking.] Biomech, 26, 633-644.
McLeod PC, Kettelkamp DB, Srinivasan SR et al.
(1975) Measurements of repetitive activities of
the knee.] Biomech, 8, 369-373.
McNevin NH, Wulf G, Carlson C (2000) Effects of
attentional focus, self-control, and dyad training
on motor learning: implications for physical
therapy. Phys Ther, 80, 373-385.
Mackey F, Ada L, Heard R et al. (1996) Stroke
rehabilitation: are highly structured units more
conducive to physical activity than less
structured units? Arch Phys Med Rehabil, 77,
Macko RF, De Souza CA, Tretter LD et al. (1997)
Treadmill aerobic exercise training reduces the
energy and cardiovascular demands of
hemiparetic gait in chronic stroke patients.
Stroke, 28, 326-330.
Magill RA (1998) Motor Learning Concepts
and Applications. McGraw-Hill, New York,
5th edn.
Magill RA (2001) Motor Learning Concepts
and Applications. McGraw-Hill, New York,
6th edn.
Mahoney FI, Barthel DW (1965) Functional
evaluation: the Barthel Index. Maryland
Med], 14,61-65.
Maki BE, McIlroy WE (1997) The role of limb
movements in maintaining upright stance: the
"change-in-support" strategy. Phys Ther, 77,
Maki BE, McIlroy WE (1998) Control of
compensatory stepping reactions: age-related
impairment and the potential for remedial
intervention. Physiother Theory Pract, 15,
Malezic M, Kljajic M, Acimovic-Jamezic et al.
(1987) Therapeutic effects of multi-site electric
stimulation of gait in motor-disabled patients.
Arch Phys Med Rehabil, 68, 553-560.
Malouin F (1995) Observational gait analysis.
In Gait Analysis. Theory and Applications
(eds RL Craik, CA Oatis), St Louis, Mosby,
pp 112-124.
Malouin F, Potvin M, Prevost .I et a1. (1992) Use
of an intensive task-oriented gait training
program in a series of patients with acute
cerebrovascular accidents. Phys Ther, 72,
Malouin F, Picard L, Bonneau C et al. (1994)
Evaluating motor recovery early after stroke:
comparison of the Fugl-Meyer Assessment and
the Motor Assessment Scale. Arch Phys Med
Rehabil, 75, 1206-1212.
Malouin F, Bonneau C, Pichard L et a1. (1997)
Non-reflex mediated changes in plantarflexor
muscles early after stroke. Scand] Rehabil Med,
Mark VW, Kooistra CA, Heilmann KM (1988)
Hemispatial neglect affected by non-neglected
stimuli. Neurology, 38,1207-1211.
Martiniuk RB, Leavitt JL, MacKenzie CL et a1.
(1990) Functional relationships between grasp
and transport components in a prehension task.
Human Mov Sci, 9,149-176.
MassionJ (1992) Movement, posture, and
equilibrium: interaction and coordination. Prog
Neurobiol, 38, 35-56.
Mathiowetz V (1990) Effects of three trials on grip
and pinch strength measurements. ] Hand Ther,
Oct-Dec, 195-198.
Mathiowetz V, Wade MG (1995) Task constraints
and functional motor performance of
individuals with and without multiple sclerosis.
Ecolog Psych, 7, 99-123.
Mathiowetz V, Weber K, Kashman N et a1.
(1985) Adult norms for the nine-hole peg
test of finger dexterity. Occup Ther] Res,
Mauritz K-H (1990) General rehabilitation. Curr
Opin Neural Neurosurg, 3, 714-718.
Mazzeo RS, Cavanagh P, Evans WJ et a1. (1998)
ACSM Position Stand: exercise and physical
activity for older adults. Med Sci Sports
Exercise, 30,992-1008.
Means KM, Rodell DE, O'Sullivan PS (1996) Use
of an obstacle course to assess balance and
mobility in the elderly. Am] Phys Med Rehabil,
Mecagni C, Smith JP, Roberts KE et a1. (2000)
Balance and ankle range of motion in
community-dwelling women aged 64-87
years: a correlational study. Phys Ther,
80, 1004-1011.
Medical Research Council (1976) Aid to the
Examination of the Peripheral Nervous System.
HMSO, London.
Mercer VS, Sahrmann SA (1999) Postural synergies
associated with a stepping task. Phys Ther, 79,
Mercier C, Bourbonnais D, Bilodeau Set al. (1999)
Description of a new motor re-education
programme for the paretic lower limb aimed at
improving the mobility of stroke patients. Clin
Rehabil, 13, 199-206.
Mikesky AE, Topp R, Wigglesworth JK et al.
(1994) Efficacy of home-based training program
for older adults using elastic tubing. Eur] Appl
Phvsiol, 69, 316-320.
Milcz;rek JJ, Lee Kirby R, Harrison ER et al.
(1993) Standard and four-footed canes: their
effect on the standing balance of patients with
hemiparesis. Arch Phys Med Rehabil, 74,
Miller GJT, Light KE (1997) Strength training in
spastic hemiparesis: should it be avoided?
NeuroRehabil, 9, 17-28.
Millington PI, Myklebust BM, Shambes GM
(1992) Biomechanical analysis of the
sit-to-stand motion in elderly persons. Arch
Phys Med Rehabil, 73, 609-617.
Miltner W, Bauder H, Sommer M et al. (1999)
Effects of constraint-induced movement
therapy on patients with chronic motor
deficits after stroke: a replication. Stroke, 30,
Mizrahi J, Solzi P, Ring H et al. (1989) Postural
stability in stroke patients: vectorial expression
of asymmetry, sway activity and relative
sequence of reactive forces. Med & Bioi Eng &
Mngoma NF, Culham EG, Bagg SD (1990)
Resistance to passive shoulder external rotation
in persons with hemiplegia: evaluation of an
assessment system. Arch Phys Med Rehabil, 80,
Moritani T (1993) Neuromuscular adaptations
during the acquisition of muscle strength,
power and motor tasks. ] Biomech, 26,
Morris DM, Crago JE, DeLuca SC et al. (1997)
Constraint-induced movement therapy for
motor recovery after stroke. NeuroRehabil, 9,
Morrissey MC, Harman ES, Johnson M] (1995)
Resistance training modes: specificity and
effectiveness. Med Sci Sports Exercise, 27,
Moseley A (1997) The effect of casting combined
with stretching on passive ankle dorsiflexion in
adults with traumatic head injuries. Phys Ther,
77, 240-247.
Moseley A, Adams R (1991) Measurement of
passive ankle dorsiflexion: procedure and
reliability. Aust ] Physiother, 37, 175-18l.
Moseley A, Dean C (2000) Treadmill training with
body weight support after stroke. Aust Synapse,
]uly, 5-6.
Mountcastle VB, Lynch ]C, Georgopoulos A et al.
(1975) Posterior parietal association cortex of
the monkey: command functions for operations
with extrapersonal space.] Neurophysiol, 38,
Mudie MH, Matyas TA (1996) Upper extremity
retraining following stroke: effects of practice.
] Neurol Rehabil, 10, 167-184.
Mudie MH, Matyas TA (2000) Can simultaneous
bilateral movement involve the undamaged
hemisphere in reconstruction of neural networks
damaged by stroke? Disabil Rehabil, 22, 23-37.
Munton JS, Ellis MI, Chamberlain MA et al.
(1981) An investigation into the problems of
easy chairs used by the arthritic and the elderly.
Rheumatol Rehabil, 20, 164-173.
Murphy AJ, Spinks WL (2000) The importance of
movement specificity in isokinetic assessment.
] Human Mov Stud, 38, 167-183.
Murray MP (1967) Gait as a total pattern of
movement. Am] Phys Med, 46, 290-333.
Murray MP, Spurr GB, Sepic SB et al. (1985)
Treadmill vs floor walking; kinematics,
electromyogram, and heart rate. ] Appl Physiol,
Myers AM, Fletcher PC, Myers AH et al. (1998)
Discriminative and evaluative properties of the
Activities-specific Balance Confidence (ABC)
Scale. ] Gerontol Med Sci, 53A, M287-294.
Nakamura R, Hosokawa T, Tsuji I (1985)
Relationship of muscle strength for knee
extension to walking capacity in patients with

spastic hemiparesis. TohokuJ Exp Med, 145, !

Nakayama H, Jorgensen HS, Raaschou HO
et al. (1994) Recovery of upper extremity
function in stroke patients: the Copenhagen
Stroke Study. Arch Phys Med Rehabil, 75,
Napier ]R (1956) The prehensile movement of the
human hand.] Bone]t Surg, 38, 902-913.
Nardone A, Schieppati M (1988) Postural
adjustments associated with voluntary
contraction of leg muscles in standing man. Exp
Brain Res, 69, 469-480.
Nashner LM (1982) Adaptation of human
movement to altered environments. Trends
Neurosci, 5, 358-36l.
Nashner LM (1983) Analysis of movement control
in man using the movable platform. In Motor
Control Mechanisms in Health and Disease
(ed JE Desmedt), Raven Press, New York,
pp 607-619.
Nashner NL, McCollum G (1985) The
organization of human postural movements:
a formal basis and experimental synthesis.
Behav Brain Sci, 8, 135-172.
Neilson JF, Sinkjaer TA (1996) A comparison of
clinical and laboratory measures of spasticity.
Multiple Sclerosis, 1,296-301.
Nelles G, Spiekmann G, Markus Jet al. (1999)
Reorganization of sensory and motor systems in
hemiplegia stroke patients: a positron
emission tomography study. Stroke, 30,
Nelles G, Jentzen W, Jueptner M et al. (2001)
Arm training induced plasticity in stroke studied
with serial positron emission tomography.
Neurolmage, 13,1146-1154.
Newham OJ, Hsiao S-F (2001) Knee muscle
isometric strength, voluntary activation and
antagonist co-contraction in the first six months
after stroke. Disabil Rehabil, 23, 379-386.
Ng S, Shepherd RB (2000) Weakness in patients
with stroke: implications for strength training in
neurorehabilitation. Phys Ther Rev, 5,
Niam S, Cheung W, Sullivan PE et al. (1999)
Balance and physical impairments after stroke.
Arch Phys Med Rehabil, 80, 1227-1233.
Nudo RJ, Friel KM (1999) Cortical plasticity after
stroke: implications for rehabilitation. Rev
Neural, 9, 713-717.
Nudo RJ, Milliken GW (1996) Reorganization of
movement representation in primary motor
cortex following focal ischemia infarcts in adult
squirrel monkeys. ] Neurophysiol, 75,
Nudo RJ, Wise BM, SiFuentes F et al. (1996)
Neural substrates for the effects of rehabilitation
on motor recovery after ischemic infarct.
)( Science, 272,1791-1794.
ft"Nudo Rj, Plautz Ej, Frost SB (2001) Role of
adaptive plasticity in recovery of function after
damage to motor cortex. Muscle Nerve, 8,
Nugent j, Schurr K, Adams R (1994) A dose
response relationship between amount of
weight-bearing exercise and walking outcome
following cerebrovascular accident. Arch Phys
Med Rehabil, 75,399-402.
Odeen I (1981) Reduction of muscular hypertonus
by long-term muscle stretch. ScandJ Rehabil
Med, 13, 93-99.
O'Dwyer NJ, Ada L, Neilson PD (1996) Spasticity
and muscle contracture following stroke. Brain,
Ohlsson A-L, Johansson BB (1995) Environment
influences functional outcome of cerebral
infarction in rats. Stroke, 26, 644-649.
Olney SJ, Monga TN, Costigan PA (1986)
Mechanical energy of walking of stroke
patients. Arch Phys Med Rehabil, 67, 92-98.
Olney SJ, Richards C (1996) Hemiparetic gait
following stroke. Part 1: characteristics. Gait
Posture, 4, 136-148.
Otis jC, Root L, Kroll MA (1985) Measurement of
plantarflexor spasticity during treatment with
tone-reducing casts. ] Pediatr Orthop, 5,
Oxendine A (1984) Psychology of Motor Learning.
Prentice-Hall, Englewood Cliffs, NJ, 2nd edn.
Page Sj, Levine P, Sisto Set al. (2001) A
randomized efficacy and feasibility study of
imagery in acute stroke. Clin Rehabil, 15,
Pai Y-C, Patton j (1997) Center of mass velocity
position predictions for balance control.
] Biomech, 30, 347-354.
Pai Y, Rogers MW (1990) Control of body
mass transfer as a function of speed of ascent
in sit-to-stand. Med Sci Sports Exercise, 22,
Pai Y, Rogers MW (1991) Segmental contributions
to total body momentum in
sit-to-stand. Med Sci Sports Exercise, 23,
Pai Y-C, Rogers MW, Hedman LD et al. (1994)
Alterations in weight-transfer capabilities in
adults with hemiparesis. Phys Ther, 74,
Palmitier RA, An KN, Scott SG et al. (1991)
Kinetic chain exercise in knee rehabilitation.
Sports Med, 11,402-413.
Pandyan AD, Johnson GR, Price ClM et al. (1999)
A review of the properties and limitations of the
Ashworth and modified Ashworth Scales as
measures of spasticity. Clin Rehabil, 13,
Paolucci S, Grasso MG, Antonucci G et al. (2001)
Mobility status after inpatient stroke
rehabilitation: 1 year follow-up and prognostic
factors. Arch Phys Med Rehabil, 82, 2-8.
Parry RH, Lincoln NB, Vass CD (1999a) Effect of
severity of arm impairment on response to
additional physiotherapy early after stroke. Clin
Rehabil, 13, 187-198.
Parry RH, Lincoln NB, Appleyard MA (1999b)
Physiotherapy for the arm and hand after
stroke. Physiotherapy, 85,417-425.
Pascual-Leone A, Torres F (1993) Plasticity of the
sensorimotor cortex representation of the
reading finger in Braille readers. Brain, 116,
Pascual-Leone A, Wassermann EM, Sadato N
et al. (1995) The role of reading activity on the
modulation of motor cortical outputs to the
reading hand in Braille readers. Ann Neural, 38,
Pashley J (1989) Grip strengthening with adapted
computer switches. Am] Occup Ther, 43,
Patla AE (1993) Age-related changes in visually
guided locomotion over different terrains: major
issues. In Sensorimotor Impairments in the
Elderly (eds GE Stelmack, V Homberg), Kluwer,
Dordrecht, pp 231-252.
Patla AE (1995) A framework for understanding
mobility problems in the elderly. In Gait
Analysis: Theory and Application (eds RL
Craik, CA Oatis), Mosby, St Louis,
pp 436-449.
Patla AE (1997) Understanding the roles of vision
in the control of human locomotion. Gait
Posture, 5, 54-69.
Patla A, Prentice S, Robinson C et al. (1991) Visual
control of locomotor strategies for changing
direction and for going over obstacles. J Exp
Psychol Hum Percept Perform, 17, 603-634.
Patterson RM, Stegink Jansen CW, Hogan HA
et al. (2001) Material properties of Thera-Band
tubing. Phys Ther, 81, 1437-1445.
Peat M (1986) Functional anatomy of the shoulder
complex. Phys Ther, 66, 1855-1865.
Peat M, Dubo HIC, Winter DA et al. (1976)
Electromyographic temporal analysis of gait:
hemiplegic locomotion. Arch Phys Med Rehabil,
Pedersen PM, Wandel A, Jorgenson HS et al.
(1996) Ipsilateral pushing in stroke: incidence,
relation to neuropsychological symptoms, and
impact on rehabilitation. The Copenhagen
Stroke Study. Arch Phys Med Rehabil, 77,
Pennisi G, Rapisarda G, Bella R et al. (1999)
Absence of response to early transcranial
magnetic stimulation in ischaemic stroke
patients: prognostic value for hand motor
recovery. Stroke, 30, 2666-2670.
Perry J (1992) Gait Analysis. Normal and
Pathological Function. Slack, Thorofare, NJ.
Pierrot-Deseilligny E (1990) Electrophysiological
assessment of the spinal mechanisms underlying
spasticity. Electroencephalogr Clin
Neurophysiol Suppl, 41, 264-273.
Pinniger GJ, Steele JR, Thorstensson A,
Cresswell AG (2000) Tension regulation
during lengthening and shortening actions of
the human soleus muscle. Eur J Appl Physiol,
81, 375-383.
Podsialo D, Richardson S (1991) The timed 'up
and go': a test of basic functional mobility for
frail elderly persons. JAGS, 39,142-148.
Pohl M, Mehr-Olz J, Ritschel C et al. (2002)
Speed-dependent treadmill training in
ambulatory hemiparetic patients. A randomized
controlled trial. Stroke, 33, 553-558.
Pollock ML, Foster C, Knapp D et al. (1987) Effect
of age and training on aerobic capacity and
body composition of master athletes. J Appl
Physiol, 62, 725-731.
Pomeroy VM, Dean 0, Sykes L et al. (2000) The
unreliability of clinical measures of muscle tone:
implications for stroke therapy. Age Ageing, 29,
Poole JL, Whitney SL (1988) Motor Assessment
Scale for stroke patients: concurrent validity and
interrater reliability. Arch Phys Med Rehabil,
Posner MI, Walker JA, Friedrich FA et al. (1987)
How do the parietal lobes direct covert
attention? Neuropsychology, 25,135-145.
Potempa K, Lopez M, Braun LT et al. (1995)
Physiological outcomes of aerobic exercise
training in hemiparetic stroke patients. Stroke,
Potempa K, Braun LT, Tinknell T et al. (1996)
Benefits of aerobic exercise after stroke. Sports
Med, 21, 337-346.
Powell J, Pandyan AD, Granat M et al. (1999)
Electrical stimulation of wrist extensors in
poststroke hemiplegia. Stroke, 30, 1384-1389.
Powell LE, Myers AM (1993) The
Activities-specific Balance Confidence (ABC)
Scale. J Gerontol Med Sci, 50A, M28-34.
Price CIM, Pandyan AD (2001) Electrical
stimulation for preventing and treating post
stroke shoulder pain: a systematic Cochrane
review. Clin Rehabil, 15,5-19.
Pritzel M, Huston JP (1991) Unilateral ablation of
telencephalon induces appearance of
contralateral cortical and subcortical projection
to thalamic nuclei. Behalf Brain Res, 3, 43-54.
Prudham 0, Evans JG (1981) Factors associated
with falls in the elderly: a community study. Age
Ageing, 10, 141-146.
Putnam CA (1993) Sequential motions of body
segments in striking and throwing skills:
descriptions and explanations. J Biomech,
26(suppl 1), 125-135.
Rab GT (1994) Muscle. In Human Walking
(eds J Rose, JG Gamble), Williams & Wilkins,
Baltimore, MD, 2nd edn, pp 101-122.
Rasch PT, Morehouse LE (1957) Effect of static
and dynamic exercise on muscular strength and
hypertrophy. Appl Physiol, 11,29-34.
Redfern MS, Di Pasquale J (1997) Biomechanics of
descending ramps. Gait Posture, 6,119-125.
Reiter F, Danni M, Lagalla G et al. (1998) Low
dose botulinum toxin with ankle taping for the
treatment of spastic equinovarus foot after
stroke. Arch Phys Med Rehabil, 79, 532-535.
Richards CL, Malouin F, Wood-Dauphinee S
(1993) Task-specific physical therapy for
optimization of gait recovery in acute
stroke patients. Arch Phys Med Rehabil, 74,
Richards CL, Malouin F, Dean C (1999) Gait in
stroke: assessment and rehabilitation. Clin
Geriatr Med, 15, 833-855.
Riddoch MJ, Humphrey GW (1994) Towards an
understanding of neglect. In Cognitive
Neuropsychology and Cognitive Rehabilitation
(eds MJ Riddoch, GW Humphreys), Erlbaum,
London, pp 125-149.
Roach KE, Miles TP (1991) Normal hip and knee
active range of motion: the relationship to age.
Phys Ther, 71, 656-665.
Rodosky MV, Andriacchi TP, Andersson GBJ
(1989) The influence of chair height on lower
limb mechanics during rising. J Orthop Res, 7,
Roebroeck ME, Doorenbosch CAM, Harlaar J
et al. (1994) Biomechanics and muscular activity
during sit-to-stand transfer. Clin Biomech, 9,
Rogers MW, Pai Y-C (1990) Dynamic transitions
in stance phase accompanying leg flexion
movements in man. Exp Brain Res, 81,
Rogers MW, Pai Y-C (1993) Patterns of muscle
activation accompanying transition in stance
during rapid leg flexion. J Electramyogr
Kinesial, 3, 149-156.
Rogers MW, Hedman LD, Pai Y-C (1993) Kinetic
analysis of dynamic transitions in stance support
accompanying voluntary leg flexion movements
in hemiparetic subjects. Arch Phys Med Rehabil,
Rogers RL, Meyer JS, Mortel KF (1990) After
reaching retirement age physical activity sustains
cerebral perfusion and cognition. JAGS, 38,
Roland PE (1973) Lack of appreciation of
compressibility, adynamaesthesia and
akinaesthesia. J Neural Sci, 20, 51-56.
Rooney KJ, Herbert RD, Balnave RJ (1994) Fatigue
contributes to the strength training stimulus.
Med Sci Sports Exercise, 26, 1160-1164.
Rosenbaum DA, Vaughan J, Barnes HJ et al.
(1992) Time course of movement planning:
selection of handgrips for object manipulation.
J Exp Psychol: Learning, Memory, Cognition,
Rosenbaum DA, Meulenbroek RG.J, Vaughan J
et al. (1999) Coordination of reaching and
grasping by capitalizing on obstacle avoidance
and other constraints. Exp Brain Res, 128,
Rosenfalck A, Andreassen S (1980) Impaired
regulation of force and firing pattern of single
motor units in patients with spasticity. J Neural
Neurosurg Psychiatry, 43, 907-916.
Rothstein.JM (1997) It is our choice! Phys Ther,
77, 800-801.
Rutherford OM (1988) Muscular coordination and
strength training: implications for injury
rehabilitation. Sports Med, 5, 196-202.
Rutherford OM, Jones DA (1986) The role of
learning and coordination in strength training.
Eur J Appl Physiol, 55,100-105.
Sackett D, Strauss S, Richardson Wet al. (2000)
Evidence-based Medicine: How to Teach and
Practice EBM, 2nd edn. Churchill Livingstone,
Sackley CM, Lincoln NB (1997) Single blind
randomized controlled trial of visual
.feedback after stroke: effects on stance
symmetry and function. Disabil Rehabil, 19,
Sahrmann SA, Norton B.J (1977) The relationship
of voluntary movement to spasticity in the
upper motor neuron syndrome. Ann Neural, 2,
Said CM, Goldie PA, Patla AE et al. (1999)
Obstacle crossing in subjects with stroke. Arch
Phys Med Rehabil, 80, 1054-1059.
Sale DG (1987) Influence of exercise and training
on motor unit activation. In Exercise and Sports
Science Reviews 15 (ed KB Pandolf), Macmillan,
New York, pp 95-151.
Saling M, Stelmach GE, Mescheriakov S et al.
(1996) Prehension with trunk-assisted reaching.
Behav Brain Res, 80, 153-160.
Sandin K.J, Smith BS (1990) The measure of
balance in sitting in stroke rehabilitation
prognosis. Stroke, 21, 82-86.
Saravanamuthu R, Shepherd RB (1994) The
effect of foot placement on the movement
pattern of sit-to-stand in the elderly. In Second
World Congress of Biomechanics Abstracts
(eds L Blankevoort, .JGM Kooloos), Stichting
World Biomechanics, Nijmegen, p 175.
Saunders .JB, Inman VT, Eberhart HD (1953) The
major determinants in normal and pathological
gait. J Bone Jt Surg, 35A, 543-558.
Sauvage L, Myklebust B, Crow-Pan.J et al. (1992)
A clinical trial of strengthening and aerobic
exercises to improve gait and balance in elderly
male home residents. Am J Phys Med Rehabil,
71, 333-342.
Scarborough DM, Krebs DE, Harris BA (1999)
Quadriceps muscle strength and dynamic stability
in elderly persons. Gait Posture, to, 10-20.
Schenkman M, Berger RA, O'Riley P et aJ. (1990)
Whole-body movements during rising to
standing from sitting. Phys Ther, 70, 638-651.
Schiene K, Brandt C, Zilles K et al. (1996)
Neuronal hyperexcitability and reduction of
GABAA receptor expression in the surround of
cerebral photothrombosis. J Cereb Blood Flow
Metab, 16,906-914.
Schultz AB, Alexander NB, Ashton-Miller JA
(1992) Biomechanical analyses of rising from a
chair. J Biomech, 25, 1383-1391.
Seelen HAM, van Wiggen KL, Halfens .JHG et al.
(1995) Lower limb postural responses during
sit-to-stand transfer in stroke patients
during neurorehabilitation. In Book of
Abstracts of the XVth Congress of the
International Society of Biomechanics (eds
K Hakkinen, KL Keskinen, PV Komi et al.),
University of Jyvaskyla, Jyvaskyla, Finland,
pp 826-827.
Seidler RD, Stelmach GE (2000) Trunk-assisted
prehension: specification of body segments with
imposed temporal constraints. J Mot Behav, 32,
Sharp SA, Brouwer B.J (1997) Isokinetic strength
training of the hemiparetic knee: effects on
function and spasticity. Arch Phys Med Rehabil,
Shea CH, Wulf G, Whitacre C (1999) Enhancing
training efficiency and effectiveness through the
use of dyad training. ] Motor Behav, 31,
Shelbourne KD, Nitz P (1992) Accelerated
rehabilitation after anterior cruciate ligament
reconstruction. ] Orthop Sports Phys Ther, 15,
Shepherd RB (2001) Exercise and training to
optimize functional motor performance in
stroke: driving neural reorganization? Neural
Plasticity, 8, 121-129.
Shepherd RB, Gentile AM (1994) Sit-to-stand:
functional relationships between upper body
and lower limb segments. Hum Mov Sci, 13,
Shepherd RB, Koh HP (1996) Some biomechanical
consequences of varying foot placement in sit
to-stand in young women. Scand J Rehabil Med,
Sherrington C, Lord SR (1997) Home exercise to
improve strength and walking velocity after hip
fracture: a randomized controlled trial. Arch
Phys Med Rehabil, 78, 208-212.
Sherrington C, Moseley A, Herbert R et al. (2001)
Guest editorial. Physiother Theory Prac, 17,
Shiavi R (1985) Electromyographic patterns in
adult locomotion: A comprehensive review.
J Rehabil Res, 22, 85-98.
Shumway-Cook A, Gruber W, Baldwin M et al.
(1997) The effect of multidimensional exercises
on balance, mobility, and fall risk in
community-dwelling older adults. Phys Ther,
Shumway-Cook A, Brauer S, Woollacott M (2000)
Predicting the probability for falls in
community-dwelling older adults using
the Timed Up & Go test. Phys Ther, 80,
Sidaway B, Yook D, Fairweather M (2001)
Visual and verbal guidance in the learning
of a novel motor skill. ] Hum Mov Stud,
Sietsema JM, Nelson DL, Mulder RM et al. (1993)
The use of a game to promote arm reach in
persons with traumatic brain injury. Am]
Occup Ther, 47, 19-24.
Silver KHC, Macko RF, Forrester LW et al. (2000)
Effects of aerobic treadmill training on gait
velocity, cadence, and gait symmetry in chronic
hemiparetic stroke: a preliminary report.
Neurorehabil Neural Repair, 14, 65-71.
Simoneau GG, Bereda SM, Sobush DC et al. (2001)
Biomechanics of elastic resistance in therapeutic
exercise programs. J Orthop Sports Phys Ther,
31, 16-24.
Simpson JM, Salkin S (1993) Are elderly people at
risk of falling taught how to get up again? Age
Ageing, 22, 294-296.
Sims KJ, Brauer SG (2000) A rapid upward step
challenges medio-lateral postural stability. Gait
Posture, 12, 217-224.
Singer B, Dunne J, Allison G (2001) Clinical
evaluation of hypertonia in the triceps surae
muscles. Phys Ther Rev, 6, 71-80.
Singer RN, Lidor R, CauraughJH (1993) To be
aware or not aware? What to think about while
learning and performing a motor skill. Sport
Psychologist, 7, 19-30.
Sinkjaer T, Magnussen I (1994) Passive, intrinsic
and reflex-mediated stiffness in the ankle
extensors of hemiparetic patients. Brain, 117,
Skelton DA, Young A, Greig CA et al. (1995)
Effects of resistance training on strength, power,
and selected functional abilities of women aged
75 and over. ] Am Geriatr Soc, 43,1081-1087.
Skilbeck CE, Wade DT, Hewer RL, Wood VA
(1983) Recovery after stroke. ] Neurol
Neurosurg Psychiatry, 46, 5-8.
Slobounov S, Newell KM (1993) Postural
dynamics as a function of skill level and task
constraints. Gait Posture, 2, 85-93.
Sluijs EM, Kok GJ, van deer Zee J (1993)
Correlates of exercise compliance in physical
therapy. Phys Ther, 73, 771-782.
Small S, Solodkin A (1998) The neurobiology of
stroke rehabilitation. Neuroscientist, 4, 426-434.
Smeets JBJ, Brenner E (1999) A new view on
grasping. Motor Control, 3, 237-271.
Smith GV, Macko RF, Silver KHC et al. (1998)
Treadmill aerobic exercise improves quadriceps
strength in patients with chronic hemiparesis
following stroke: a preliminary report. ] Neurol
Rehab, 12, 111-118.
Smith GV, Silver KHC, Goldberg AP et al. (1999)
Task-oriented exercise improves hamstring
strength and spastic reflexes in chronic stroke
patients. Stroke, 30, 2112-2118.
Smits JG, Smits-Boone E (2000) Hand Recovery
after Stroke: Exercises and Results
Measurements. Butterworth-Heinemann,
Son K, Miller JAA, Schultz AB (1988) The
mechanical role of the trunk and lower
extremities in a seated weight-moving task in
the sagittal plane. ] Biomech Eng, 110,97-103.
Sonde L, Gip C, Fernaeus SE et al. (1998)
Stimulation with low frequency (1.7Hz)
transcutaneous electrical nerve stimulation
(low-TENS) increases motor function of the
post-stroke paretic arm. Scand J Rehabil Med,
Sonde L, Kalimo H, Fernaeus SE et al. (2000) Low
TENS treatment on post-stroke paretic arm: a
three year follow-up. Clin Rehabil, 14, 14-19.
Sorock G, Pomerantz R (1980) A case-control
study of falling episodes among hospitalized
elderly. Gerontologist, 20, 240.
Spirduso WW, Asplund LA (1995) Physical activity
and cognitive function in the elderly. Quest, 47,
Stapleton T, Ashburn A, Stack E (2001) A pilot
study of attention deficits, balance control and
falls in the acute stage following stroke. Clin
Rehabil, 15,437--444.
Stelmach GE, Worringham C (1985) Sensorimotor
deficits related to postural stability: implications
for falling in the elderly. Clin Geriatr Med, 1,
Stephenson R, Edwards S, Freeman] (1998)
Associated reactions: their value in clinical
practice? Physiother Res Int, 3, 69-75.
Stevenson T], Garland] (1996) Standing balance
during internally produced perturbations in
subjects with hemiplegia: validation of the
Balance Scale. Arch Phys Med Rehabil, 77,
Strathy GM, Chao EY, Laughman RK (1983)
Changes in knee function associated with
treadmill ambulation.] Biomech, 16, 517-522.
Stroke Unit Trialists' Collaboration (1997)
Collaborative systematic review of the
randomized ttials of organized in-patient
(stroke unit) care. BM], 314,1151-1159.
Studenski S, Duncan PW, Chandler J (1991)
Postural responses and effector factors in
persons with unexplained falls: results and
methodologic issues.] Am Geriatr Soc, 39,
Sun], Walters M, Svenson Net al. (1996) The
influence of surface slope on human gait
characteristics: a study of urban pedestrians
walking on an inclined surface. Ergonomics, 39,
Sunderland A, Tinson D, Bradley L et al. (1989)
Arm function after stroke. An evaluation of grip
strength as a measure of recovery and a
prognostic indicator. J Neural Neurasurg
Psychiatry, 52, 1267-1272.
Sunderland A, Tinson DJ, Bradley L et al. (1992)
Enhanced physical therapy improves recovery of
arm function after stroke: a randomised
controlled trial. ] Neural Neurosurg Psychiatry,
Sunderland A, Fletcher D, Bradley L et al. (1994)
Enhanced physical therapy for arm function
after stroke: a one year follow-up study.
] Neural Neurosurg Psychiatry, 57,
Sutherland DH, Kaufman KR, Moitoza]R (1994)
Kinematics of normal human walking. In
Human Walking (eds] Rose,]G Gamble),
Williams & Wilkins, Baltimore, MD, 2nd edn,
pp 23--44.
Svantesson U, Grimby G, Thomee R (1994)
Potentiation of concentric plantar flexion torque
following eccentric and isometric muscle
actions. Acta Physiol Scand, 152, 287-293.
Svantesson UM, Sunnerhagen KS, Carlsson US
et al. (1999) Development of fatigue during
repeated eccentric-concentric muscle
contractions of plantar flexors in patients
with stroke. Arch Phys Med Rehabil, 80,
Taira M, Mine S, Georgopoulos AP et al. (1990)
Parietal cortex neurons of the monkey related to
the visual guidance of hand movement. Exp
Brain Res, 83, 29-36.
Talvitie U (2000) Socio-affective characteristics and
properties of extrinsic feedback in
physiotherapy. Physiother Res Int, 5,173-188.
Tang A, Rymer WZ (1981) Abnormal force-EMG
relations in paretic limbs of hemiparetic human
subjects. J Neurol Neurasurg Psychiatry, 44,
Tangeman PT, Banaitis DA, Williams AK (1990)
Rehabilitation of chronic stroke patients:
changes in functional performance. Arch Phys
Med Rehabil, 71, 876-880.
Tardieu G, Shentoub S, Delarue R (1954) Ala
recherche d'une technique de measure de la
spasticite. Rev Neural, 91, 143-144.
Taub E (1980) Somatosensory deafferentation
research with monkey: implications for
rehabilitation medicine. In Behavioral
Psychology in Rehabilitation Medicine: Clinical
Applications (ed LP Ince), Williams & Wilkins,
New York, pp 371--401.
Taub E, Uswatte G (2000) Constraint-induced
movement therapy and massed practice. (Letter)
Stroke, 31, 986-988.
Taub E, Wolf SL (1997) Constraint induced
movement techniques to facilitate upper
extremity use in stroke patients. Top Stroke
Rehabil, 3, 38-61.
Taub E, Miller NE, Novak TA et al. (1993) A
technique for improving chronic motor deficit
after stroke. Arch Phys Med Rehabil, 74,
Tax AAM, Denier van den Gon J], Erkelens C]
(1990) Differences in coordination of elbow
flexors in force tasks and in movement. Exp
Brain Res, 81, 567-572.
Taylor DC, Dalton ]D, Seaber AV et al. (1990)
Viscoelastic properties of muscle-tendon units.
The biomechanical effects of stretching. Am]
Sports Med, 18, 300-309.
Teixeira-Salmela LF, Olney S], Nadeau S et al.
(1999) Muscle strengthening and physical
conditioning to reduce impairment and
disability in chronic stroke survivors. Arch Phys
Med Rehabil, 80, 1211-1218.
Teixeira-Salmela LF, Nadeau S, McBride I et al.
(2001) Effects of muscle strengthening and
physical conditioning training on temporal,
kinematic and kinetic variables during gait in
chronic stroke survivors. ] Rehabil Med, 33,
Telian SA, Shepard NT, Smith-Wheelock M et al.
(1990) Habituation therapy for chronic
vestibular dysfunction: preliminary results.
J Otolaryngol Head Neck Surg, 103, 89-95.
Thepaut-Mathieu C, Van Hoecke J, Maton B
(1988) Myoelectric and mechanical
changes linked to length specificity during
isometric training. J Appl Physiol, 64,
Thilmann AF, Fellows SJ (1991) The time-course
of bilateral changes in the reflex excitability of
relaxed triceps surae muscle in human
hemiparetic spasticity. J Neurol, 238, 293-298.
Thilmann AF, Fellows SJ, Garms E (1991a) The
mechanism of spastic muscle hypertonus. Brain,
Thilmann AF, Fellows SJ, Ross HF (1991b)
Biomechanical changes at the ankle joint after
stroke. J Neural Neurosurg Psychiatry, 54,
Thorstensson A, Oddsson L, Carlson H (1985)
Motor control of voluntary trunk movements
in standing. Acta Physiol Scand, 125,309-321.
Tinetti ME, Williams FT, Mayewski R (1986) Fall
risk for elderly patients based on number of
chronic disabilities. Am I Med, 80,429-434.
Tinetti ME, Speechley M, Ginter SF (1988) Risk
factors for falls among elderly persons living in
the community. N Engl J Med, 319,
Tinetti ME, Liu W, Claus EB (1993) Predictors and
prognosis of inability to get up after falls among
elderly persons. JAMA, 269,65-70.
Tinetti ME, Baker OI, McAvay G et al. (1994) A
multifactorial intervention to reduce the risk of
falling among elderly women living in the
community. N EnglJ Med, 331, 821-827.
Tinson DJ (1989) How stroke patients spend their
days. Int Disabil Stud, 11,45-49.
Tobis JS, Nayak L, Hoehler F (1981) Visual
perception of verticality and horizontality
among elderly fallers. Arch Phys Med Rehabil,
Topp R, Mikesky A, Wigglesworth Jet al. (1993)
The effect of a 12-week dynamic resistance
strength training program on gait velocity and
balance of older adults. Gerontologist, 33,
Topp R, Mikesky A, Dayhoff NE et al. (1996)
Effects of resistance training on strength,
postural control, and gait velocity among older
adults. Clin Nurs Res, 5, 407-427.
Treiber FA, Lott J, Duncan Jet al. (1998) Effects of
Theraband and lightweight dumbbell training
on shoulder rotation torque and serve
performance in college tennis players. Am
J Sports Med, 26, 510-515.
Trombly CA (1992) Deficits of reaching in subjects
with left hemiparesis: a pilot study.
Am J Occup Ther, 46,887-897.
Trombly CA (1993) Observations of improvement
of reaching in five subjects with left hemiparesis.
J Neural Neurasurg Psychiatry, 56, 40-45.
Tse S, Bailey DM (1991) T'ai Chi and postural
control in the well elderly. Am J Occup Ther,
Turnbull GI, Chateris J, Wall JC (1996)
Deficiencies in standing weight shifts by
ambulant hemiplegic subjects. Arch Phys Med
Rehabil, 77, 356-362.
Tyson S, Turner G (2000) Discharge and
follow-up for people with stroke: what happens
and why. Clin Rehabil, 14,381-392.
van der Lee JH, Wagenaar RC, Lankhorst GJ
et aJ. (1999) Forced use of the upper extremity
in chronic stroke patients: results from a
single-blind randomized clinical trial. Stroke,
van der Lee JH, De Groot V, Beckerman H et al.
(2001) The intra- and interrater reliability of the
action research arm test: a practical test of
upper extremity function in patients with stroke.
Arch Phys Med Rehabil, 82, 14-19.
Vander Linden DW, Brunt D, McCulloch MU
(1994) Variant and invariant characteristics of
the sit-to-stand task in healthy elderly adults.
Arch Phys Med Rehabil, 75, 653-660.
van der Meche FGA, van Gijn J (1986) Hypotonia:
an erroneous clinical concept? Brain, 109,
van der Ploeg RJO, Fidler V, Oosterhuis HJGH
(1991) Hand-held myometry: reference values.
J Neurol Neurosurg Psychiatry, 54, 244-247.
Vandervoort AA (1999) Ankle mobility and
postural stability. Physiother Theory Prac, 15,
Vandervoort AA, Chesworth BM, Jones M et al.
(1990) Passive ankle stiffness in young and
elderly men. Can J Aging, 9, 204-212.
Vandervoort AA, Chesworth BM, Cunningham DA
et al. (1992) Age and sex effects on mobility of
the human ankle. J Gerontal: Med Sci, 47,
van der Weel FR, van der Meer AL, Lee DN (1991)
Effect of task on movement control in cerebral
palsy: implications for assessment and therapy.
Dev Med Child Neurol, 33, 419-426.
Van Leeuwen R, Inglis JT (1998) Mental practice
and imagery: a potential role in stroke
rehabilitation. Phys Ther Rev, 3, 47-52.
van Straten A, de Haan RJ, Limburg Met al.
(1997) A stroke-adapted 3D-item version of the
Sickness Impact Profile to assess quality of life.
Stroke, 28, 2155-2161.
van Vliet P (1993) An investigation of the task
specificity of reaching: implications for
retraining. Physiother Theory Pract, 9, 69-76.
van Vliet P (1999) Reaching, pointing and using
two hands together. In Functional Human
Movement: Measurement and Analysis
(eds BR Durward, GD Baer, P] Rowe),
Butterworth-Heinemann, Oxford, pp 159-179.
van Vliet P, Kerwin DG, Sheridan M et al. (1995)
The influence of goals on the kinematics of
reaching following stroke. Neurol Rep, 19,
Vattanasilp W, Ada L (1999) The relationship
between clinical and laboratory measures of
spasticity. AustJ Physiother, 45, 135-139.
Vattanasilp W, Ada L, Crosbie] (2000)
Contribution of thixotropy, spasticity, and
contracture to ankle stiffness after stroke.
] Neurol Neurosurg Psychiatry, 69, 34-39.
Verfaelli M, Bowers D, Heilman KM (1988)
Attentional factors in the occurrence of
stimulus-response compatibility effects.
Neuropsychology, 26, 435-444.
Verkerk PH, Schouten ]P, Oosterhuis H]GH
(1990) Measurement of the hand coordination.
Clin Neural Neurosurg, 92,105-109.
Viallet F, Massion], Massarino Ret al. (1992)
Coordination between posture and movement in
a bimanual load-lifting task: putative role of a
medial frontal region including the
supplementary motor area. Exp Brain Res, 88,
Visintin M, Barbeau H, Korner-Bitensky N et al.
(1998) A new approach to retrain gait in
stroke patients through body weight support
and treadmill stimulation. Stroke, 29,
von Schroeder HP, Coutts RD, Lyden PD et al.
(1995) Gait parameters following stroke: a
practical assessment. ] Rehabil Res De!, 32,
Wade DT (1992) Measurement in Stroke
Rehabilitation. Oxford University Press,
Wade DT, Langton-Hewer R, Wood VA et al.
(1983) The hemiplegic arm after stroke:
measurement and recovery. ] Neurol Neurosurg
Psychiatry, 46, 521-524.
Wade DT, Wood VA, Langton Hewer R (1985)
Recovery after stroke - the first three months.
j Neurol Neurosurg Psychiatry, 48, 7-13.
Wade DT, Langton Hewer R, David RM et al.
(1986) Aphasia after stroke: natural history and
associated deficits. j Neurol Neurosurg
Psychiatry, 49, 11-16.
Wade DT, Wood VA, Heller A et al. (1987)
Walking after stroke. Scand J Rehabil Med, 19,
Wagenaar RC (1990) Functional Recovery after
Stroke. VU University Press, Amsterdam.
Wagenaar RC, Beek W] (1992) Hemiplegic gait:
a kinematic analysis using walking speed as a
basis. j Biomech, 25,1007-1015.
Wajswelner H, Webb G (1995) Therapeutic
exercise. In Sports Physiotherapy Applied
Science and Practice (eds M Zuluaga, C Briggs,
] Carlisle et al.), Churchill-Livingstone,
Melbourne, pp 208-215.
Walker C, Brouwer B], Culham EG (2000) Use of
visual feedback in retraining balance following
acute stroke. Phys Ther, 80, 886-895.
Wall]C (1999) Walking. In Functional Human
Movement: Measurement and Analysis
(eds BR Durward, GD Baer, P] Rowe),
Butterworth-Heinemann, Oxford,
pp 93-105.
Wall ]C, Nottrody]W, Charteris] (1981) The
effects of uphill and downhill walking on pelvic
oscillations in the transverse plane. Ergonomics,
Walsh EG (1992) Postural thixotropy: a significant
factor in the stiffness of paralysed limbs?
Paraplegia, 30, 113-115.
Walsh RN, Cummins RA (1975) Mechanisms
mediating the production of environmentally
induced brain changes. Psychol Bull, 82,
Walshe F (1923) On certain tonic or postural
reflexes in hemiplegia, with special reference to
the so-called 'associated movements'. Brain,
46, 1-37.
Wanklyn P, Forster A, Young] (1996) Hemiplegic
shoulder pain (HSP): natural history and
investigation of associated features. Disabil
Rehabil, 18,497-501.
Weiller C, Chollet F, Friston K] et al. (1992)
Functional reorganization of the brain in
recovery from striatocapsular infarction in man.
Ann Neurol, 31, 463-472.
Weiner DK, Bongiorni DR, Studenski SA et al.
(1993) Does functional reach improve with
rehabilitation? Arch Phys Med Rehabil, 74,
Weiss A, Suzuki T, Bean] et al. (2000) High
intensity strength training improves strength and
functional performance after stroke. Am j Phys
Med Rehabil, 79, 369-376.
Wernick-Robinson M, Krebs DE, Giorgetti MM
(1999) Functional reach: does it really measure
dynamic balance? Arch Phys Med Rehabil, 80,
Westing SH, Seger ]Y, Thorstensson A et al. (1990)
Effects of electrical stimulation on eccentric and
concentric torque-velocity relationship during
knee extension in man. Acta Physiol Scand, 140,
Wheeler], Woodward C, Ucovich RL et al. (1985)
Rising from a chair. Influence of age and chair
design. Phys Ther, 65, 22-26.
Whipple RH, Wolfson LI, Amerman PM (1987)
The relationship of knee and ankle weakness to
falls in nursing home residents: an isokinetic
study. JAGS, 35, 13-20.
Whitall], Waller SM, Silver KHC et al. (2000)
Repetitive bilateral arm training with rhythmic
auditory cueing improves motor function in
chronic hemiparetic stroke. Stroke, 31,
Whiting HTA (1980) Dimensions of control in
motor learning. Tutorials in Motor Behavior
(eds GE Stelmach, J Requin), North Holland,
New York, pp 537-550.
Wiesendanger M, Kazennikov 0, Perrig S et al.
(1996) Two hands-one action. In Hand and
Brain. Neurophysiology and Psychology o(
Hand Movements (eds AM Wing, P Haggard, R
Flanagan), Academic Press, Orlando, FL,
pp 283-300.
Will B, Kelche C (1992) Environmental
approaches to recovery of function from brain
damage: a review of animal studies. In
Recovery (ram Brain Damage: Reflections and
Directions (eds FD Rose, DA Johnson), Plenum
Press, New York, pp 79-104.
Williams PE (1990) Use of intermittent stretch in
the prevention of serial sarcomere loss in
immobilised muscle. Ann Rheum Dis, 49,
Williams PE, Catanese T, Lucey EG et al. (1988)
The importance of stretch and contractile
activity in the prevention of connective
tissue accumulation in muscle. ] Anat, 158,
Wilson B, Cockburn J, Halligan P (1987)
Development of a behavioural test of
visuospatial neglect. Arch Phys Med Rehabil,
Wing AM, Fraser C (1983) The contribution of the
thumb to reaching movements. Quart] Exp
Psychol, 35, 297-309.
Wing AM, Turton A, Fraser C (1986) Grasp size
and accuracy of approach in reaching. ] Motor
Behav, 18,245-260.
Wing AM, Lough S, Turton A et al. (1990)
Recovery of elbow function in voluntary
positioning of the hand following hemiplegia
due to stroke. ] Neural Neurosurg Psychiatry,
53, 126-134.
Wing AM, Goodrich S, Virji-Babul Net al. (1993)
Balance evaluation in hemiparetic stroke
patients using lateral forces applied to the hip.
Arch Phys Med Rehabil, 74, 292-299.
Winstein CJ, Gardner ER, McNeal DR et al.
(1989) Standing balance training: effect on
balance and locomotion in hemiplegic adults.
Arch Phys Med Rehabil, 70, 755-762.
Winter DA (1980) Overall principle of lower limb
support during stance phase of gait. ] Biomech,
Winter DA (1983) Biomechanical motor patterns
in normal walking. ] Motor Behav, 15,
Winter DA (1985) Concerning the scientific basis
for the diagnosis of pathological gait and for
rehabilitation protocols. Physiother Can, 37,
Winter DA (1987) The Biomechanics and Motor
Control o( Human Gait, University of Waterloo
Press, Waterloo, Ontario.
Winter DA (1991) Biomechanics and Motor
Control o( Human Gait: Normal, Elderly and
Pathological. University of Waterloo Press,
Waterloo, Ontario.
Winter DA (1995) A.B.C. o( Balance during
Standing and Walking. Waterloo Biomechanics,
Waterloo, Ontario.
Winter DA, Patla AE, Frank JS et al. (1990)
Biomechanical walking pattern changes in the fit
and healthy elderly. Phys Ther, 70, 340-347.
Winter DA, McFadyen BJ, Dickey JP (1991)
Adaptability of the CNS in human walking. In
Adaptability o( Human Gait (ed AE Patla),
Elsevier Science, Amsterdam, pp 127-143.
Winter DA, Deathe AB, Halliday Set al. (1993)
A technique to analyse the kinetics and
energetics of cane-assisted gait. Clin Biomech,
Winters JM, Kleweno DG (1993) Effect of initial
upper-limb alignment on muscle contributions
to isometric strength curves. ] Biomech, 26,
Winward CE, Halligan PW, Wade DT (1999)
Somatosensory assessment after central nerve
damage: the need for standardized clinical
measures. Phys Ther Rev, 4, 21-28.
Wolf SL, Catlin PA, Blanton S et al. (1994)
Overcoming limitations in elbow movement in
the presence of antagonist hyperactivity. Phys
Ther, 74, 826-835.
Wolf SL, Barnhart HX, Ellison et al. (1997) The
effect of Tai Chi Quan and computerized
balance training in older subjects. Phys Ther,
Wolfson L, Whipple R, Judge Jet al. (1993)
Training balance and strength in the elderly to
improve function. ] Am Geriatr Soc, 41,
Wong AM, Lin Y, Chou S et al. (2001)
Coordination exercise and postural stability in
elderly people: effect of Tai Chi Chuan. Arch
Phys Med Rehabil, 82, 608-612.
Woollacott M, Shumway-Cook A, Nashner LM
(1986) Aging and posture control: changes in
sensory organization and muscular
coordination. Tnt] Aging Hum Dev, 23,
Woollacott M, lnglin B, Manchester D (1988)
Response preparation and posture control. Ann
NY Acad Sci, 525, 42-53.
Worrell TW, Borchert B, Erner K (1993) Effect of a
lateral step-up exercise protocol on quadriceps
and lower extremity performance. ] Orthop
Sports Phys Ther, 187, 646-653.
Wu C, Trombly CA, Lin K et al. (2000) A
kinematic study of contextual effects on
reaching performance in persons with and
without stroke: influences of object availability.
Arch Phys Med Rehabil, 81, 95-101.
Wulf G, McNevin N, Shea C (1999a) Learning
phenomena: future challenges for the dynamical
systems approach to understanding the learning
of complex motor skills. IntJ Psychol, 30,
Wulf G, Lauterbach B, Toole T (1999b) The
learning advantage of an external focus of
attention in golf. Res Quart Exercise Sport, 70,
Wynn-Parry CB, Salter M (1975) Sensory
reeducation after median nerve lesions. Hand, 8,
Yang JF, Winter DA, Wells RP (1990) Postural
dynamics in the standing human. Bioi Cybern,
62, 309-320.
Yekutiel M, Guttman E (1993) A controlled trial of
the retraining of the sensory function of the
hand in stroke patients. } Neural Neurosurg
Psychiatry, 56,241-244.
Yoshida K, Iwakura H, Inoue F (1983) Motion
analysis in the movements of standing up from
and sitting down on a chair. ScandJ Rehabil
Med, 15,133-140.
Yu DT, Chae J, Walker ME et al. (2001)
Percutaneous intramuscular neuromuscular
electric stimulation for the treatment of shoulder
subluxation and pain in patients with chronic
hemiplegia: a pilot study. Arch Phys Med
Rehabil, 82,20-25.
Yue G, Cole KJ (1992) Strength increases from the
motor program: a comparison of
training with maximal voluntary and imagined
muscle contractions. } Neurophysiol, 67,
Zajac FE (1989) Muscle and tendon: properties,
models, scaling, and application to
biomechanics and motor control. Crit Rev
Biomed Eng, 17, 359-41l.
Zatsiorsky VM, Li ZM, Latash ML (1998)
Coordinated force production in
multi-finger tasks: finger interaction and
neural network modelling. Bioi Cybern, 79,
Zatsiorsky VM, Li Z, Latash ML (2000) Enslaving
effects in multi-finger force production. Exp
Brain Res, 131, 187-195.
Zattara M, Bouisset S (1988) Chronometric
analysis of the posturo-kinetic programming
of voluntary movement. } Motor Behav, 18,
Zorowitz RD (2001) Recovery patterns of
shoulder subluxation after stroke: a
six-month follow-up study. Top Stroke
Rehabil, 8, 1-9.
Page references in italic refer to figures and those in bold refer to tables.
Abstract tasks, 18, 172
Action Research Arm Test (ARA), 205
Activities of daily life, testing, 205
Activities Specific Balance Confidence (ABC)
Scale, 73, 74
Actual Amount of Use Test (AAUT), 205
Adaptations to motor impairments, 209,
and balance, 48, 49, 61
classification, 210-11
standing up, 145
upper limbs, 173-5
see also Spatiotemporal adaptations
Aerobic exercise, 88, 119, 254-8
Age-related change
balance, 45-8
exercise, 257, 258
sitting down, 138
standing up, 138-9
strength training, 235, 245, 246, 257
walking, 87-8
Agnosia, spatial, 228-30, 263-4
Aids, walking, 127-8
Amount of Use Scale (AOU), 205
balance, 35
exercises, 62, 66, 68
falls, 46
flexibility, 46, 48
range of motion measurement, 123,221,
sitting down, 138, 150
standing up
biomechanics, 130,131,133-4, 135, 136
observational analysis, 143, 144
training, 146, 147, 152-3
analysis, 89, 90, 92, 93-8
ankle-foot orthoses, 126-7
biomechanics, 81, 83, 84, 85-6, 87
degrees of freedom, 77
exercises, 113, 115, 116,245
falls, 88
slopes, 86
stairs, 87
Ankle-foot orthoses (AFOs), 126-7
Anosognosia, 228, 229
Aphasias, 262
age-related change, 45-6
sitting, 42, 43, 57-60

spatiotemporal adaptations, 53
standing, 38, 39
walking, 44
bimanual training, 190-2
excessive muscle activity, 217, 218
persistent posturing, 219
pulling patient by, 194,263
reaching, 159-60
adaptations, 173-5
bimanual training, 190
biomechanics, 161-5
eliciting muscle activity, 180-6,204
motor performance analysis, 171-7
recovery after stroke, 168-71
soft tissue stretching, 179-80
tests, 205
training, 57-60, 168-9, 177-9, 180-6,
and vision, 227
while sitting, 42, 43, 57-60
redundant muscle activity, 219
shoulder pain, 194, 196, 197, 263
sitting balance, 42, 43, 57-60
standing balance, 38, 39
standing up, 130, 139, 144
supports for, 197, 198
walking, 44, 81
Arm Motor Mobility Test (AMAT), 205
Ashworth Scale, 216, 219, 220-1
Associated movements (synkinesis), 218-19
Astereognosis, 225, 226
and balance, 36
focusing, 16-18, 226
spatial neglect, 229-30, 263-4
Auditory feedback, 157-8
Autotopagnosia, 228, 229
Balance, 35-7
age-related change, 45-8
biomechanical tests, 38, 49, 75
biomechanics, 38-45, 49-53, 78, 79, 93
body alignment analysis, 52-3
central nervous system role, 36
functional tests, 73-5
motor performance analysis, 48-54
need for training, 37
observational analysis, 50-3
postural sway, 38, 46-7, 49, 75
pusher syndrome, 230-1
responses to loss of, 41, 44-5, 47, 48
Balance (contd)
sensory inputs, 36, 43, 45, 47, 48,50
spatiotemporal adaptations, 52-3, 54
task contexts, 52-3
training, 55-6
acute phase, 259-60, 261,263
fall prevention, 47,55,119
outcome measurement, 73-5
sitting, 56-60, 70-1, 259-60, 261, 263
skill maximization, 70-1, 72
soft tissue stretching, 71
standing, 60-71, 72, 259-60,261
strength, 71-3, 74, 214, 245
supports, 55, 61,63,64,66
transfer of learning, 45, 55
visual inputs, 36,43, 45, 50
biomechanics, 42-4, 78, 79, 93
fall avoidance, 77, 119
functional tests, 74-5
obstacles, 44, 50,71, 119, 121
training, 66, 67
Balance reactions, 37
Barthel Index, 26
Berg Balance Scale, 73
Bicycle ergometers, 254, 256, 257
Biofeedback, 181-2,204,244
Biomechanical tests
balance, 38,49, 75
sitting down, 157
standing up, 157
upper limb function, 205, 206
walking, 123
Biomechanics, 3
balance, 38-45, 49-53, 78, 79, 93
sitting down, 44-5, 138
standing up, 44-5, 130-8
upper limb functions, 161-8, 171-7
walking, 42-4, 78-87, 89-99
see also Biomechanical tests
Body alignment, analysis, 52-3
Botulinum toxin, 126,220,251-2
Brain reorganization
functional, 5-8, 169-71, 198
measurement, 24-7
rehabilitation environment, 8-24
sensory training, 227
Brunnstrom stages, upper limb recovery,
Calf muscle length test, 123, 253
Canes, walking, 127-8
Cardiovascular fitness, 253-8
Chairs, 135-6, 137, 150
Circuit training, 10, 13,27,28-31
balance, 72
strength, 247-8
walking, 103
Closed-chain exercises, 237-40, 243
Cognitive impairments, 48, 50, 228-32,
Communication deficits, 262
Community-based support groups, 266
Computerized training
balance, 67, 70
upper limb, 183,202,203
Concentric exercise, 238-9, 240-1
Concrete tasks, 18, 172
Conditioning, physical, 253-8
Constraint-induced movement therapy, 168,
Cutaneous inputs
balance, 36, 48
hand use, 167
practice, 226
Cycle ergometers, 254, 256, 257
Demonstrating actions, 16, 17
Dexterity, 159
critical components, 175-7
loss of, 214
tests, 205
training, 169, 188-90,214
Discharge planning, 265-6
Downhill walking, 86
Dynamometry, 213, 249
joint range, 252
standing up, 157
upper limb function, 205
walking, 108, 117, 118-19
Dysphagia, 261
Dystonia, 219, 220
Eccentric exercise, 238-9, 240-1
Elastic band resistance exercises, 117, 119,
Electrical stimulation (ES), 244
upper limb function, 178, 181-2,204
walking, 108, 125-6
Electromyography (EMC)
excessive muscle activity, 217
standing up, 136, 138
upper limb function, 181-2,204
walking, 85-6, 91
Environment, for rehabilitation
see Rehabilitation environment
Equilibrium reactions, 37
Evidence-based rehabilitation, 24-5
Exercise testing, 257-8
Exercise tolerance, 14,253
and balance, 37
age-related change, 46-8
exercises, 47, 71
functional tests, 74
responses to loss of, 41, 44-5, 47, 48
strength training, 47, 71, 73
walking, 77, 119
frequency, 48-9
getting up after, 49, 71, 73
risk factors, 37, 41, 46, 47, 49,88,250
when standing up, 129, 130
when walking, 77, 88, 119
Falls Efficacy Scale, 73, 74
augmented, 19
sitting down, 157-8
standing up, 157-8
walking, 108
Feedback (contd)
balance training, 67, 70
eliciting muscle activity, 108, 181-2,204,
rehabilitation environment, 14, 18-20
adaptive movements, 174, 175
grasp biomechanics, 162, 163, 165-7
motor performance analysis, 172, 174,
sensory perception, 173
training, 177
eliciting muscle activity, 181-2, 204
manipulation, 188,189,190
strength, 246
stretching, 179
Fitness tests, 122
Follow-up services, 265-6
balance, 36
age-related change, 46, 47
single leg support, 65
sitting, 42, 57, 58, 59
soft tissue stretching, 71
spatiotemporal adaptations, 53
standing, 39-41, 60, 62, 63
walking, 43-4, 48
sitting balance, 42, 57, 58, 59
sitting down, 150-2
standing balance, 39-41, 60, 62, 63
standing up
biomechanics, 131, 133-4
eliciting muscle activity, 152-3
observational analysis, 143, 144, 145
soft tissue stretching, 150, 151
training, 145-7, 150, 152-3, 154
analysis, 89,92,93,95-9
ankle-foot orthoses, 126-7
and balance, 43-4, 48
biomechanics, 78, 79, 80, 81, 82, 86-7
exercises, 113, 115
stairs, 87
Forced use training, 168, 169-70, 178, 198,
Frames, walking, 126, 127
Fugl-Meyer Scale, 171,221
Functional brain reorganization, 5-8
constraint-induced movement therapy, 198
rehabilitation environment, 8-24
upper limb, 169-71, 198
Functional electrical stimulation (FES), 204
Functional Independence Measure (FIM), 26
Functional outcomes, 26-7
aerobic exercise, 257-8
balance training, 73-5
sitting down training, 141-2,157
standing up training, 141-2, 157
upper limb function, 199-201,205,206
walking training, 102-3, 122, 123-5
Functional Reach (FR) Test, 73, 74
Gait see Walking
balance training, 71
computer, 183,202,203
Goal setting, 18
Goniometers, 252
Grasp, 159
adaptive movements, 174, 175
biomechanics, 161-8
motor performance analysis, 172, 174
sensory perception, 173,227
tests for, 205
critical components, 175-7
eliciting muscle activity, 182-3, 184
manipulation, 188, 189, 190
soft tissue stretching, 179-80
strength, 192-3,246
Grip force test, 205
Group training, 10-13
Hand, 159-60
adaptive movements, 174, 175
bimanual tasks, 160, 163, 177, 178-9,
brain reorganization, 169
function tests, 205
grasp biomechanics, 161-8
grip force regulation, 173
motor performance analysis, 172, 174,
recovery after stroke, 168-71
sensation-motor output relation, 225-6
sensory function, 160-1, 169, 173,205,
shaping, 167
and standing up, 139, 149
training, 160, 163, 166-8, 175-7
bimanual, 190-2
cutlery use, 192
eliciting muscle activity, 180-6
exercise tasks, 188-90
strength, 192, 193,214,246
wrist malalignment, 175, 222
Harness support
balance training, 55, 61, 64, 66
reaching exercises, 12,63,64
walking training, 101, 104-6, 124-5
balance, 35
body alignment, 53
sitting, 42, 56-7
standing, 46-7,62,63
training, 56-7, 62
walking, 44
standing up, 130, 147
Heart rate test, 258
Heel raises, 115-16, 117
Hemiplegia, denial of, 228, 229
balance, 35
body alignment, 53
sitting, 42, 53
soft tissue stretching, 71
spatiotemporal adaptations, 53, 54
standing, 40, 53, 62, 63, 66, 67-8
walking, 44
sitting balance, 42, 53
sitting down, 138, 150
standing balance, 40, 53, 62, 63, 66, 67-8
Hip (contd)
standing up
age-related change, 139
biomechanics, 130, 131, 133, 134-5,
motor performance analysis, 139-40,
143, 144
training, 146, 150, 153
analysis, 89, 90, 92, 94-8
biomechanics, 44,81-2,83,84,85-6,87
degrees of freedom, 77
exercises, 108-11, 113, 115,245
slopes, 86
stairs, 87
Human Activity Profile (HAP), 26, 245
Hyperreflexia, 127,209,210,215-16,
Hypertonia, 209-10, 215, 216-17, 220, 221,
Hypotonia, 217
Imagery (mental practice), 23-4
Incontinence, urinary, 261-2
Instructing patients, 16-18, 17
Isokinetic dynamometry
muscle strength measurement, 157,205,
walking, 108, 117, 118-19
Isokinetic exercise, 237, 240, 243
walking, 108, 117, 118-19
Isometric exercise, 237
Isotonic exercise, 237-40, 241, 243
Joint malalignment, 175,222
Joint range measurement, 123,205,221,252
Kerbs, 86-7, 106
standing up, 131-3
walking, 80-2, 86, 90, 91-9
Kinetic chain exercises, 237-40
standing up, 131-3
walking, 82-4, 90
sitting, 42, 53
standing, 60-1, 66, 68
training supports, 61
walking, 44
muscle weakness differentiation, 213
muscle weakness distribution, 212
sitting balance, 42, 53
sitting down, 138, 150
standing balance, 60-1, 66, 68
standing up
age-related change, 139
biomechanics, 130, 131, 133, 135, 136,
motor performance analysis, 139-40,
143, 144
timed tests, 157
training, 147, 149, 150
analysis, 89, 90, 91, 92, 93-8
balance, 44
biomechanics, 81, 82, 83, 84, 85-6, 87
degrees of freedom, 77
exercises, 113, 115,245
falls, 88
slopes, 86
stairs, 87
Language deficits, 262
Lateral Step-up Test, 213
Lateropulsion, 230-1
Learning, environment for, 15-24
Leg press exercises, 238
Lifting methods, 262-3, 263
Lower limb
balance, 35, 37
age-related change, 45-6, 47
body alignment, 53
sitting, 42, 43, 50, 51, 53, 57-60
sitting down, 44-5
spatiotemporal adaptations, 53, 54
standing, 38-41, 49-50, 53,60-71,72
standing up, 44-5
training, 37, 47, 57-71,72,74,214,245
walking, 43-4, 66, 67, 93
exercises see Lower limb exercises
muscle weakness differentiation, 213
muscle weakness distribution, 211-12
risk factors for falls, 46, 47, 88
sitting balance, 42, 43, 50, 51, 53, 57-60
sitting down, 138, 150, 154-6
standing balance, 38-41, 49-50, 53, 60-71,
standing up
age-related change, 139
balance, 44-5
biomechanics, 130, 131-4, 135, 136, 138
functional tests, 157
motor performance analysis, 139-40,
training, 146, 147-56
analysis, 89-99
balance, 43-4, 66, 67, 93
biomechanics, 43-4, 78-81, 82-4, 85-7
degrees of freedom, 77
falls, 88
training, 99-121,123-5,214,245-6,
Lower limb exercises
balance training, 37,47,60-2
outcomes, 74
reaching while sitting, 57-60, 74
reaching while standing, 62-4, 66-7,
68-9, 74
sideways walking, 66, 67
single leg support, 65, 66
skill maximization, 70-1, 72
soft tissue stretching, 71
strength, 71-3, 74,214
circuit class, 28-31, 247-8
pedal machines, 12
sitting down, 150, 154-6
standing up, 146, 147-56
Lower limb exercises (contd)
strength training, 237-40, 243, 245-6,
balance, 71-3, 74, 214
walking, 111-19,214,245-6
task modification, 22
walking, 99-106,123-5
eliciting muscle activity, 108-11, 125-6
skill maximization, 119
soft tissue stretching, 106-8,251
strength training, 111-19,214,245-6
Manipulation, 159-61
adaptive movements, 174, 175
biomechanics, 161-8
grip force regulation, 173
motor performance analysis, 172, 174, 175-7
recovery after stroke, 168-71
sensation-motor output relation, 225
tests of, 205
training, 160, 163, 168-71, 175-8
bimanual tasks, 190-2
eliciting muscle activity, 180-6
exercise tasks, 188-90
strength, 192, 193
Measuring outcomes see Outcome
Mechanical power, walking, 83-4,90
Medical Research Council, Motricity Index,
Mental practice, 23-4
Modelling actions, 16, 17
Moments of force, walking, 82, 83, 90
Motivation, 13-14, 18,19,20,23,247-8
Motor Activity Log, 205
Motor Assessment Scale, 73, 204, 205
Motor impairments, 209
adaptations to see Adaptations to motor
classification, 210-11, 210
dexterity, 214
muscle weakness see Muscle weakness
spasticity, 209-10, 215-21, 224
aerobic exercise, 254
ankle-foot orthoses, 127
strength training, 220, 234
treadmill training, 123
walking dysfunction, 77
Motor patterns, adaptive, 223-4
Motor performance
and sensory function, 225-6
transfer of learning, 20-1
aerobic exercise, 253-4
balance, 45, 55
strength training, 21, 241-4, 245, 249
Motor performance analysis, 25-6
balance, 48-54
sitting down, 140-5
standing up, 139-45
upper limb function, 171-7
walking, 88-99
Motor skills, taxonomy of, 20-1, 52
Motriciry Index, 213,244
Muscle activity
adaptive changes, 222
associated movements, 218-19
balance, 35
age-related change, 45-6, 47
eliciting, 61
motor performance analysis, 48, 49, 50
sitting, 42, 59
sitting down, 44-5
standing, 38-41, 60-1, 68
standing up, 44-5
walking, 44
and body position, 243
and dexterity, 214
eliciting, 244
balance, 61
mental practice, 23-4
standing up, 152-4
upper limb, 178, 180-6,204
walking, 108-11, 125-6
excessive, 217-19
hand function, 165-7, 168, 171, 180-6,
persistent limb posturing, 219
redundant, 217, 218-19
sitting balance, 42, 59
sitting down, 44-5, 138
standing balance, 38-41, 60-1, 68
standing up, 44-5,131-3,136-8,152-4
upper limb function
biomechanics, 165-7
motor performance analysis, 171-3
recovery after stroke, 168, 171
spasticity, 173
training, 176, 177, 178, 180-6,204
walking, 77, 84-6
analysis, 91, 93-4
balance during, 44
degrees of freedom, 76-7
eliciting, 108-11, 125-6
mechanical power, 83-4, 90
stairs, 87
support moment of force, 83
Muscle adaptations, 221-4
Muscle co-contraction, 217-18
Muscle contracture, 223, 250-2
see also Soft tissue stretching
Muscle coordination, dexterity, 214
Muscle fibre changes, 222, 223, 253, 254
Muscle flexibility, 222-3, 249-52
and aerobic exercise, 254
see also Soft tissue stretching
Muscle length, 223, 249-52
passive lengthening see Hypertonia
see also Soft tissue stretching
Muscle spasticity see Spasticity
Muscle stiffness, 222-3, 250-2
and aerobic exercise, 254
see also Soft tissue stretching
Muscle strength measurement, 205, 213,244,
Muscle weakness, 211-13, 215, 234
age-related, 45-6, 47
balance, 45-6, 47, 49, 50, 59, 61, 68
classification, 210
and dexterity, 214
effects, 235
eliciting activity see Muscle activity, eliciting
from disuse, 222, 224, 236
improvement see Strength training
Muscle weakness (contd)
measures, 205, 213, 244, 249
sitting down, 150
standing up, 139,143,145,146,147-50
upper limb function
dexterity, 214
eliciting activity, 180-6
motor performance analysis, 171-2, 173,
shoulder pain, 194, 195, 197
soft tissue stretching, 179, 180
walking, 77, 88-9, 90, 91, 100
Neglect, spatial, 228-30, 263--4
Nine-hole Peg Test (NHPT), 205
Nottingham Health Profile, 26, 245
Observational analysis
balance, 50-3
sitting down, 140-5
standing up, 140-5
upper limb function, 173-5
walking, 91-9
Obstacle Course Test, 73, 75
Obstacles, 44, 50, 71, 86-7, 119
practising, 119,120,121
Open-chain exercises, 237--40, 243
Orthoses, walking, 126-7
Outcome measurement, 24-7
aerobic exercise, 255, 257-8
balance training, 73-5
sitting down training, 141-2, 156-7
standing up training, 141-2, 156-7
upper limb function, 199-201,205,206
walking training, 102-3, 121-5
Parallel bars, 127, 128
Passive movement, resistance to (hypertonia),
Passive stretching, 220, 224, 250, 251, 252
sitting balance, 42
standing up, 131, 138, 143
walking, 44,81-2,92,94,96,98
Pendulum test, 216, 220-1
Perceptual-eognitive impairments, 48, 50,
Persistent limb posturing, 219
Physical conditioning, 253-8
Physiological tests, 123, 257-8
Physiotherapists, role of, 264
Pick-up exercises
balance training, 66-7, 68-9, 70-1, 72
manipulation practice, 188-9
Post-discharge stroke management, 265-6
PosturaI reflexes, 37
Postural sway, 38, 46-7, 49, 75
Postural system see Balance
Posturing, persistent limb, 219
Power, and strength, 246
Practice, 9-10, 14, 21--4
repetitive, 23, 243, 246-7, 249
Proprioceptive inputs, 161, 173,205,227
Pusher syndrome, 230-1
Quality of life measures, 26, 245
Ramps, 86,106,116,118
Reaching, 159-61
adaptive movements, 173-5
and halance
functional test, 73, 74
sitting, 41-2, 43, 50, 51, 56-60, 74
skill maximization, 70-1, 72
spatiotemporal adaptations, 53, 54
standing, 53, 54, 62--4, 66-7, 68-9, 74
bimanual training, 190
biomechanics, 161-5
circuit class, 13
critical components, 175-7
indirectly supervised practice, 12
motor performance analysis, 171-7
recovery after stroke, 168-71
tests, 205
training, 160, 175-7, 178, 180-8, 190
and vision, 227
Reflex hyperactivity (hyperreflexia), 127,209,
Rehabilitation environment, 8-9, 264-5
acute phase, 259-60
aims, 264-5
demonstrating actions, 16, 17
exercise tolerance, 14
feedhack, 14,18-20
focusing attention, 16-18
goal setting, 18
group training, 10-13
instructing patients, 16-18, 17
intervention delivery, 10-15
making mistakes, 14, 19
mental practice, 23-4
motivation, 13-14, 18, 19,20,23,247-8
patient involvement, 13-14,20,264
physiotherapist's role, 264
practising, 9-10,14,21--4
recommendations, 265
repetition, 23
skill optimization, 15-24
stroke units, 259
sub-tasks, 22
tailored programmes, 14
task modification, 22
therapeutic relationship, 20
transfer of learning, 20-1
vision, 20
Repetitive practice, 23, 243, 246-7, 249
Resistance exercises, 117, 118-19, 241
and spasticity, 173
upper limb function, 173, 192, 193
Resistance to passive movement (hypertonia),
Right-sided focus, 263--4
Science-based rehahilitation, 24-5
Seat heights, 135-6, 137,146, 150,154
Self-efficacy, measuring, 26-7
Self-help groups, 265-6
Sensory function
hand, 160-1,169,173,205,225-6,227
training, 226-7
Sensory impairments, 224-32
Sensory inputs

hand use, 160-1, 167
testing, 232
walking, 77
balance, 53, 54
eliciting muscle activity, 184-6, 204
motor performance analysis, 171, 172, 173,
pain prevention, 194-7,204,263
range of motion tests, 205
reaching, 159,173,174
softtissue stretching, 179, 180
standing up, 143, 149
strength training, 192-3
subluxation, 195, 197,204
walking, 81
Sickness Impact Profile (SIP), 26
Sit to stand see Standing up
Sitting balance
biomechanics, 41-2
body alignment, 53
computerized feedback training, 67, 70
motor performance analysis, 50, 51,52-3
training, 55, 56-60, 70-1,259-60,261,
Sitting down, 129, 130
age-related change, 138
balance, 44-5, 73
biomechanical tests, 157
biomechanics, 44-5, 138, 140-5
functional tests, 73, 157
motor performance analysis, 140-5
observational analysis, 140-5
training, 145-6, 150
feedback, 157-8
groups, /1,12
outcomes, 141-2, 156-7
skill maximization, 155-6, 249
soft tissue stretching, 150-2, 151
strength, 154-5,249
Skill acquisition, 15-24
Slide sheets, 262-3, 263
Slings, 197, 198,202,203
Slopes, 86,106,116,118
Soft tissue stretching, 250
arm, 179-80
balance, 71
hand, 179-80
passive, 220, 224, 250, 251, 252
prescription, 252
sitting down, 150-2, 151
standing up, 150-2, 151
walking, 88, 106-8,251
Somatosensory impairments, 48, 50, 225-7
Spasticity, 209-10, 215-21, 224
aerobic exercise, 254
ankle-foot orthoses, 127
strength training, 220, 234
treadmill training, 123
upper limb function, 172-3, 217, 218-19
Spatial agnosia, 228-30, 263-4
Spatiotemporal adaptations
balance, 52-3, 54
walking, 100
Spatiotemporal variables
grasp biomechanics, 163
walking, 80, 89, 90, 91, 100, 122
Speech disorders, 262
balance, 54
reaching movements, 173, 174
standing up, 131, 136, 143, 144,150
walking, 81, 82
Spiral test, 205
Splints, 61, 202, 203, 252
Stair walking, 29, 87
Stand to sit see Sitting down
Standing balance
age-related change, 45-7
asymmetrical, 61
biomechanics, 38-41
body alignment, 53
compensatory stepping, 41, 47, 48
computerized feedback training, 67, 70
motor performance analysis, 48-54
response to loss of, 41, 47
spatiotemporal adaptations, 53, 54
training, 55, 60-71, 72, 74, 214, 259-60
Standing up, 129
age-related change, 138-9
balance, 44-5, 47, 73, 74
biomechanical tests, 157
biomechanics, 44-5, 130-8, 139-45
chair type, 135-6, 137, 150
falls during, 47, 129, 130
foot placement, 133-4
functional tests, 73, 74, 157
motor performance analysis, 139-45
muscle activity, 44-5, 131-3, 136-8,
observational analysis, 140-5
seat height, 135-6, 137, 146,150, 154
speed of movement, 135
training, 145-50
circuit class, 28
eliciting muscle activity, 152-4
feedback, 157-8
focusing attention, 17
groups, /1, 12
outcomes, 141-2, 156-7
skill maximization, 155-6, 249
soft tissue stretching, 150-2, 151
strength, 154-5,249
trunk rotation speedltiming, 134-5
Step test, 122
for balance, 41, 47, 48
over obstacles, 44, 50, 71, 86-7, 119,
Stepping exercises, 238, 239
balance, 71, 72, 74
group training, 11
walking, 112-15,112, /14,116
Steps, 86-7
Sticks, walking, 127-8
Strength training, 233-7
balance, 37, 47, 61, 71-3, 74,214,245
effects, 235-7, 240, 241-6